CN107095393A - A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology - Google Patents
A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology Download PDFInfo
- Publication number
- CN107095393A CN107095393A CN201710175716.9A CN201710175716A CN107095393A CN 107095393 A CN107095393 A CN 107095393A CN 201710175716 A CN201710175716 A CN 201710175716A CN 107095393 A CN107095393 A CN 107095393A
- Authority
- CN
- China
- Prior art keywords
- foot
- image
- data
- shoes
- true
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- A—HUMAN NECESSITIES
- A43—FOOTWEAR
- A43D—MACHINES, TOOLS, EQUIPMENT OR METHODS FOR MANUFACTURING OR REPAIRING FOOTWEAR
- A43D1/00—Foot or last measuring devices; Measuring devices for shoe parts
- A43D1/02—Foot-measuring devices
- A43D1/022—Foot-measuring devices involving making footprints or permanent moulds of the foot
-
- A—HUMAN NECESSITIES
- A43—FOOTWEAR
- A43D—MACHINES, TOOLS, EQUIPMENT OR METHODS FOR MANUFACTURING OR REPAIRING FOOTWEAR
- A43D39/00—Machines for making foot-supporting pads or instep-raisers for flat feet
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Biophysics (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
Abstract
The present invention relates to a kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology, wherein:The preparation method of customization footwear includes:Gather foot image file;Utilize network collection source data;Image recognition;Three-dimensional modeling;Shoes are made according to foot threedimensional model.The manufacturing system of customization footwear includes data acquisition device, Data Collection and collects device, pattern recognition device, three-dimensional modeling apparatus, shoemaking device.The present invention disclosure satisfy that the foot health demand of user, and user operates time saving, convenient, shoemaking efficiency high, it is possible to achieve foot is orthopedic.
Description
Technical field
A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology, belong to shoe-making field.
Term " the interior bottoms of shoes ", refer to the position that vola is stepped on, i.e. foot fit inside after plantar contact position;Footwear
The interior bottom of son both includes the plane (or curved surface) contacted with vola, also including certain thickness portion under the plane (or curved surface)
Position;Particularly, the interior bottom of shoes can include shoe-pad.
Term " vamp on the inside of ", is commonly called as footwear " inside ", i.e., the position that sufficient top (including toes, instep, carpus etc.) contacts.
It is as follows that term " footwear adhesion process " includes flow:Fabric, lining severing-system side (including piece side, flanging, seam side
Deng)-bottom processed (including various bed material integers, processing, upper stretching etc.)-close bottom (side bear building-up is closed)-drying and shaping-and go out last carving-inspection
Deng.
Term " footwear hot vulcanization process " includes heat cure barbola work and heat cure mould pressing process, technique main flow
It is:Upper of a shoe manufacture-glue component manufacture-shaping-vulcanization-delasting.
Term " injection (penetrating) technique " includes whole footwear injection molding process (upper of a shoe is separately made) and sole Shooting Technique, sole note
Modeling technique can be divided into monochromatic Shooting Technique and multi-color injection molded technique again.Shooting Technique main flow is first in injection molding apparatus
By material compression, (solid plastic is transformed into fluid), homogenizing are then plastified, then by the injection canal of mould by fluid plastic
It is injected into the die cavity of footwear, the product being molded after cooling.
It is as follows that term " mould pressing process " includes flow:Stretch tight the upper of a shoe helped, and by fluffing, transfers to former last carving, coated with adhesive
After processing, be enclosed on moulding press same model aluminum last, be then put into sizing material in bed die, then by moulding press heat hot melt and to
Lower pressure, last sizing material hot melt, compacting are closely synthesized with upper of a shoe simultaneously.
Term " sewing tech " basic characteristics are that side bear building-up conjunction is sewed with special linen thread, and sewing tech is more multiple
Miscellaneous, technical requirements are higher, fully demonstrate a kind of technological ability and skill of people.
Background technology
The universal present situation of current shoe industry, is that the foot type of people is carried out after rough approximation processing, according to footwear code batch production
The consistent shoes of footwear type, i.e. the shoes with money typically only exist the difference of footwear code.The foot type that this shoemaking mode ignores people is poor
It is different, it is impossible to meet the individual demand of individual, can not also meet special population (such as Plantar Fasciitis, nurse tumour, pain at metatarsal,
Neuralgia, diabetes, interior eight foot, outer eight foot, the rear strephexopodia, flat foot, talipes cavus, rear strephenopodia patient) extraordinary demand.
In this regard, a kind of ideal solution is that there is provided comply fully with foot type according to everyone difference of sufficient type
Shoes.Include a kind of ancient customization footwear mode in this scheme, i.e.,:Foot type-customized shoe-last-basis of accurate measurement individual
Shoe tree shoemaking.This mode carries out operation using manual mode from ancient times, has the disadvantage time-consuming effort and expensive, it is difficult to general
Consumer receives.With the development of modern age footwear art, a series of similar new shoe-making process of thinkings is occurred in that, is mainly included
Following several classes:
With a kind of footwear art that Zhejiang University " personalized shoe tree mold designing method " (CN01130405.7) is representative,
I.e.:Spatial digitizer obtains three-dimensional data-standard of comparison shoe tree-adjustment shoe tree-making shoes of people's pin, and this mode is limited to three
Dimension scanner can not be popularized and used, and due to that must make shoe tree, manufacture craft is longer cause to take longer, price still compared with
It is expensive.The patents such as CN200710068032.5 also describe the technology of similar thinking.
It is the one of representative with Nike Inc " system for being used for custom fitting shoes " (CN201110117887.9)
Footwear art is planted, using a variety of governor motions such as adjustable last, fastening system, freely adjusting for shoes shape is realized, lacked
It is that governor motion is more, regulation program is more and is difficult to simplify to fall into, and making a pair of shoes needs to expend user's more time.
The patents such as CN201110117936.95 also describe the technology of similar thinking.
SOLS companies of the U.S. develop a kind of brand-new manufacturing method of shoes and system, using image recognition -3D printing technique system
Make shoes, advantage is the voluntarily collection foot image file information that people can stay indoors.US14341632、US14677894、
US14961769, US14877361, US14877171 etc. use the technology.Its deficiency is:(1) view data changes into three
During dimension module, partial data defect can only be carried out more by the result of calculation of set formula (calculation basis is pin length, body weight etc.)
Mend, even and if its described calculation formula have statistical significance, a kind of processing of rough approximation is still fallen within after all, it is clear that
Actual sufficient type difference that can be with individual;(2) mode for obtaining data is only image recognition, without essence during whole shoemaking
True sole pressure test process, it is impossible to propose counte-rplan for accurate pressure data;(3) manufacturing method of shoes emphasis considers footwear
Bottom manufacturing technology, the consideration to vamp is less;(4) it is limited to current 3D printing efficiency, shoes fabrication cycle is longer;(5) it is limited to
Current 3D printing material, shoes cost is higher.
Sum it up, in current footwear art, do not find temporarily it is a kind of can be while taking into account the scheme of following key element:(1)
Shoes with sufficient type completely or approximate match;(2) foot health demand is met;(3) fabrication cycle is short;(4) user operation it is time saving,
It is convenient;(5) low manufacture cost;(6) shoes comfortableness is high;(7) shoes can be with orthopedic;(8) foot image file, pressure point are considered
Cloth data, obtain complete foot data;(9) user individual (such as attractive in appearance) demand is met.
The content of the invention
Technology contents as described below be to the present invention simplification summary description, i.e., there is provided contribute to understand the present invention certain
The description of a little aspects.All technology contents of the completely careful explanation present invention of these descriptions, it is not intended to mark this hair
Bright all crucial or important elements, also specially scope of the claims is not defined.
It is an object of the invention to provide a kind of customization footwear preparation method and system for meeting user health demand.
Therefore, a kind of according to an aspect of the invention, there is provided customization based on image recognition and dimensional Modeling Technology
Footwear preparation method, including:
(1) foot image file is gathered:Obtain the source datas such as foot image file data;
(2) network collection foot data are utilized:Using cable network or wireless network by foot Data Collection to data
Processing platform;
(3) image recognition:Image is sequentially restored using computer, strengthens, split, changes into and is available for three
Tie up the secondary data of modeling
(4) three-dimensional modeling:Contours extract, model calculating, texture calculating, modeling rendering and wash with watercolours are carried out using secondary data
Dye, forms consistent with true foot or close to consistent foot threedimensional model;
(5) shoes are made:Shoes are made according to foot threedimensional model, wherein:True foot is put on after shoes, shoes it is interior
Fitted completely with true vola (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent) at bottom.
It is preferred that, the minimum air line distance < of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.)
0.48cm;
It is preferred that, two foot image files are at least included in source data;
It is preferred that, collection foot image file includes any combination of one or more kinds of modes in the following manner:Using taking the photograph
Image device shoots the foot static images that each foot forms more than 3;Shooting foot formation using camera includes foot
The video of overall image;Complete foot 3-D view is gathered using spatial digitizer;Gathered using electromagnetic impulse telemetry
Complete foot 3-D view;Complete foot 3-D view is gathered using running safety method;Utilize mechanical wave impulse
Telemetry gathers complete foot 3-D view;Complete foot 3-D view is gathered using mechanical wave phase ranging method;
It is preferred that, it is that the foot image file that will be collected is transferred to number by special mobile phone A PP using network collection source data
According to processing platform;Or the network interconnection will be used between the collecting device of foot image file and terminal data processing platform, number is gathered every time
According to rear, collecting device it is automatic or under manual command by foot data transfer to the network terminal;
It is preferred that, the foot image file that source data is directly obtained using image capture device, and by plantar nervous arch
The vola view data that data are changed into;
It is preferred that, according to foot threedimensional model, in the following ways in one or more kinds of modes any combination system
Make shoes:Shoes are directly printed using 3D printing technique;Make completely the same or close consistent with true foot size and shape
Shoe tree, then make shoes using footwear adhesion process;Make completely the same or close to one with true foot size and shape
The shoe tree of cause, then makes shoes using footwear hot vulcanization process;Make completely the same or connect with true foot size and shape
Nearly consistent shoe tree, then makes shoes using injection (penetrating) technique;Make with true foot size and shape it is completely the same or
Close to consistent shoe tree, then shoes are made using mould pressing process;Make completely the same or connect with true foot size and shape
Nearly consistent shoe tree, then makes shoes using sewing tech;Make completely the same or close with true foot size and shape
Consistent shoe-pad, shoe-pad is put into footwear shoes is made;According to foot threedimensional model, without making shoe tree, visual determination is utilized
(including naked eyes judge and computer vision judges) make shoes;
It is preferred that, image is restored using computer, comprised the following steps:First, image conversion is carried out, by source number
Every foot image file in is transformed into the digitized image of space coordinate and the equal discretization of gray scale;Secondly, image is removed to make an uproar
Sound, that is, utilize probability density function statistics Gaussian noise, rayleigh noise, gamma noise, exponential distribution noise, uniformly distributed noise
After impulsive noise, pre-filtering is carried out to noise image and obtains noise variance, the high frequency characteristics further according to noise is filtered using average
Ripple method or median filtering method realize LPF;Again, image degradation system is set up, i.e., with mathematical function relationship g (x, y)=h
(x, y) × f (x, y)+n (x, y) expresses degradation model, and wherein x, y represents the two-dimensional coordinate point of digitized image, and g (x, y) is represented
Degraded image, h (x, y) represents degenrate function, and f (x, y) represents digitized static two dimensional image, and n (x, y) represents extraneous noise
Effect, then with digital modeling method confirmation degenrate function h (x, y) computational methods;Finally, according to h (x, y) result of calculation, make an uproar
The removal situation of sound variance and noise, is inversely derived to image degradation model, is finally completely eliminated or part abatement image
Space degenerate (fuzzy) and put degeneration (noise), obtain restored image;
It is preferred that, image is strengthened using computer, includes times of one or more kinds of modes in the following manner
Meaning combination:Gray Classification is carried out to foot image file, the gray level more than number of pixels spreaded using histogram equalization method, right
The few gray scale of number of pixels is compressed, and forms image local high-contrast;To foot image file carry out Gray Classification, then by each
Different grey-scale is transformed into different colours according to mapping function linearly or nonlinearly, makes the details of digitlization foot image file more
Plus substantially;Digitized image is subjected to Fourier transform, different colors are then transformed into by the mapping function of frequency, make numeral
The details for changing foot image file is more obvious;It is that parameter sets up foot HSV space mould with tone (H), saturation degree (S), lightness (V)
Type, the rgb space for digitizing foot image file is mapped on HSV space model, makes the details of digitlization foot image file brighter
It is aobvious;With cyan (C), carmetta (M), yellow (Y), black (K) are that parameter sets up foot CMYK spatial models, will digitize foot
The rgb space of image is mapped on CMYK spatial models, makes the details of digitlization foot image file more obvious;It is red with illumination (L)
Color is to the brightness range (a) of green, and the brightness range (b) of blueness to yellow sets up foot Lab space model for parameter, by numeral
The rgb space for changing foot image file is mapped on Lab space model, makes the details of digitlization foot image file more obvious;
It is preferred that, image is split using computer, both including being mutated according to gray scale, color change and texture mutation
Judge the discontinuity feature in image, and then extract the border of foot and shooting background in image, then foot image file from
Separated in shooting background, any combination of one or more kinds of processes in following process is included again:According to the ash of image
Degree divides the image into the subregion of non-overlapping copies, makes all subregion can be used in building threedimensional model;According to the shape of image
The subregion of non-overlapping copies is divided the image into, makes all subregion can be used in building threedimensional model;Will according to the texture of image
Image is divided into the subregion of non-overlapping copies, makes all subregion can be used in building threedimensional model;It will scheme according to the color of image
Subregion as being divided into non-overlapping copies, makes all subregion can be used in building threedimensional model;
It is preferred that, after the completion of image transformation step, image is compressed according to Joint Photographic Experts Group;
It is preferred that, digital modeling method confirms degenrate function h (x, y) computational methods, comprises the following steps:H (x, y) is entered
H (x, y) is obtained after row Fourier transform, expression formula H (x, y)=2.72^ (- ρ β (x2+y2) are utilized5/6) to H (x, y) assignment, then
Carry out obtaining h (x, y) last solution to the H (x, y) after assignment after reverse Fourier transform, wherein ρ represents digitized image office
Number of greyscale levels in portion's unit area, β represents the local picture element density of digitized image;
It is preferred that, using computer to image carry out it is enhanced during, to color, whether distortion is not specially required;
It is preferred that, histogram equalization method is to divide the image into some regions, with histogram equalization by region to image
Carry out computing and travel through all regions;Or some regions are divided the image into, each histogram equalization computing at least covers 2
Individual region;
It is preferred that, by using the Sobel operators or Roberts operators or granny rag Laplacian operater or Preveit of gray scale
The single order such as operator or Kirsch operators or Second Order Differential Operator detection discontinuity feature;
It is preferred that, collection foot image file includes any combination of one or more kinds of modes in the following manner:Foot is worn
On be equipped with the wearable device (such as socks) of sensor, foot image file is calculated by the deformation data for reading foot diverse location
Data;Foot is put on after the figuratum wearable device (such as socks) of print, and the pattern on wearable device is deformed upon, by clapping
Take the photograph the wearable device deformation pattern no less than 2 and obtain foot data;Foot is wrapped up with deformable material, it is deformable
The built-in sensor for being capable of mark position, foot image file data are calculated by reading sensing station and displacement in material;Will
Foot is put into liquid or gas, the sensor of mark position is capable of built in liquid, gas, by reading sensing station and position
Shifting calculates foot image file data;
It is preferred that, three-dimensional feature extracts mark, is to find the obvious characteristic point or profile in picture point cloud, then at other
Corresponding points are searched in point cloud and matching relationship is set up, and it is extracted or mark mode includes the one or more side in the following manner
Any combination of formula:Extract or mark foot image file in sufficient edge, toes, heel, ankle, arch of foot etc. have notable gray scale,
The picture point of chromatic characteristic, corresponding points are searched in other clouds and matching relationship is set up;Extract curvature, the method in foot curved surface
The invariant features such as vector, the normal vector angle of point-to-point transmission, corresponding points are searched in other clouds and matching relationship is set up;Mark foot
1 points of distance relation two-by-two in portion's image, corresponding points are searched in other clouds and matching relationship is set up;Treat spelling
The two panels point cloud or wherein a piece of cloud connect is sampled, and it is right on two panels point cloud then to be made using icp algorithms by iterative calculation
The mean square error that should put to or correspond to point-to-plane distance is minimum, and then determines matching relationship;With the curvature of curved surface be characterized a little, with
Angle, distance are constraints, and point cloud matching relation is set up using 3 rotary process between a cloud;
It is preferred that, Coordinate Conversion and point cloud, are that the two-dimensional coordinate value of image data is converted into D coordinates value, and
Multiple images are matched using characteristic point or profile as binding site, the step of forming preliminary three-dimensional model, its Coordinate Conversion is pressed
Sequencing includes:Transformation of scale, that is, adjusting the proportionate relationship of each image is consistent the engineer's scale of all images;Two dimension is sat
Mark is converted into three-dimensional coordinate;
It is preferred that, include any combination of one or more modes in following optimal way:Denoising, patch formation model defect
(filling-up hole), segmentation topography, picture structure, stereo-picture surface are smooth;
It is preferred that, curvature is Gaussian curvature or average curvature;
It is preferred that, three-dimensional feature point is extracted or mark is, it is necessary to reject bad point matching;
It is preferred that, two panels point cloud to be spliced or wherein a piece of cloud are sampled, including uniform sampling method, adopts at random
Sample method, normal vector sampling method;
It is preferred that, after iterative calculation is finished, unreliable or incompatible correspondence is removed using the way of restraint for icp algorithms
Point pair or point face, its way of restraint is a kind of the way of restraint based on curved surface features, that is, constrains curvature, normal direction in foot curved surface
The invariant features such as amount, the normal vector angle of point-to-point transmission;
It is preferred that, using Solidworks, Pro/E, UG, CAD, Adobe Fuse CC, 3Ds Max, SketchUp,
Other 3 d modeling softwares of Revit or any realize three-dimensional modeling;
It is preferred that, inserted in the interior bottom of shoes pressure sensor, and/or acceleration transducer, and/or positioner and/
Or alarming device, and/or wireless communication apparatus, and/or power supply and charging device, and/or temperature regulating device;
It is preferred that, the interior bottom of shoes is detachable apparatus, it is preferable that the interior bottom of shoes is shoe-pad, it is preferable that below shoe-pad
If headspace, pad can be inserted in headspace according to the difference in height of human body double lower limb, human body is got on one's shoes can after son
Ensure balance, can be used for human body and increase;
It is preferred that, (i.e. footwear of being fitted completely with true foot top (including toes, instep, heel, carpus etc.) on the inside of vamp
The inner side of son is corresponding consistent with the shape on true sufficient top);
It is preferred that, fixed additional materials are set up above the interior bottom of shoes, make the interior bottom and true vola of shoes incomplete
Laminating, for correcting problem foot (eight foots, outer eight foot, the rear strephexopodia, flat foot, talipes cavus, rear strephenopodia in such as) or being used for
Foot affliction (such as Plantar Fasciitis, nurse tumour, pain, neuralgia, diabetes at metatarsal) is alleviated in treatment;
It is preferred that, 0.5cm projection is no more than above the interior bottom of shoes provided with multiple height, massage effect can be played,
Preferably, additional materials are corresponding with the acupuncture point of sole portion or whole;
It is preferred that, user is used after shoes, Consumer's Experience is fed back into data processing platform (DPP), then according to user's request pair
Footwear construction is adjusted;
It is preferred that, multiple closed individual cavities are provided with the interior bottom of shoes, shoes quality can be mitigated, can also be in human body
Increase cushioning effect during strenuous exercise;
It is preferred that, the interior bottom of shoes is made using wear-resisting, waterproof, with ductility and elastic material, is preferably true
Skin, rubber, polyurethanes (PU), vinyl acetate copolymer (EVA), MD, TPR, TPUE (TPU),
Polyvinyl chloride (PVC), TR, imitation leather bottom, BPU, thermoplastic elastic rubber-sole, makrolon (PC) resin, acrylonitrile-butadiene-benzene
Ethylene copolymer (ABS), butadiene-styrene rubber (SBR), and their arbitrary composition;
It is preferred that, the interior bottom of shoes can be sandwich construction, it is preferable that each layer in interior bottom of shoes uses different material systems
Into;
It is preferred that, height heels are no more than 4cm, preferably 1cm;
It is preferred that, foot threedimensional model is corrected, when foot threedimensional model is amplified to true foot size with
Untill true foot is no more than 0.5mm compared to the image deflects maximum gauge at any place, preferably 0.1mm;
It is preferred that, defect recognition is carried out to inner sole using machine vision, when defects with diameters is more than 1mm in footwear, used
Cutting, and/or injection, and/or bonding, and/or 3D printing method amendment sole defect;
It is preferred that, three-dimensional modeling uses CCD optical modeling techniques;
It is preferred that, when obtaining foot image file using filming apparatus, also include foot office using the foot data of network collection
Portion or the length information of whole;
It is preferred that, sensor of the contact surface provided with multiple detectable deformation or pressure change of shoes and foot can detect
Go out gait change or foot hyperblastosis;
It is preferred that, the tensile strength at bottom is not less than 8.5kgf/mm in shoes2, it is preferred that the tensile strength at bottom is not in shoes
Less than 10kgf/mm2;
It is preferred that, the bending modulus at bottom is not less than 140kgf/mm in shoes2, it is preferred that the bending modulus at bottom is not in shoes
Less than 180kgf/mm2;
It is preferred that, the tensile strength at bottom is not less than 8.5kgf/mm in shoes2, it is preferred that the tensile strength at bottom is not in shoes
Less than 10kgf/mm2;
It is preferred that, the bending modulus at bottom is not less than 140kgf/mm in shoes2, it is preferred that the bending modulus at bottom is not in shoes
Less than 180kgf/mm2;
There is provided a kind of customization footwear system based on image recognition and dimensional Modeling Technology according to another aspect of the present invention
Make system, including:
Data acquisition device, its have collection foot image file equipment, also with to Data Collection collect device transmission number
According to port;
Data Collection collects device, and it has port, the data storage device that data are received from data acquisition device, also has
Oriented pattern recognition device transmits the port of data;
Pattern recognition device, it, which has, collects device from Data Collection and receives the ports of data, image restoration device, image
Intensifier, image segmenting device;Also there is the port that data are transmitted to three-dimensional modeling apparatus;
Three-dimensional modeling apparatus, it has the port from pattern recognition device reception data, three-dimensional feature extraction element, coordinate
Conversion equipment, point cloud device, model global optimization device, also with the port that data are transmitted to shoemaking device;
Shoemaking device, it has from the port of three-dimensional modeling apparatus reception data, high-precision vamp making apparatus, high accuracy
In sole making apparatus, the shoes that the shoemaking device is produced bottom fitted completely with true vola (i.e. the interior bottom of shoes with it is true
The shape correspondence at full bottom is consistent), any point and sufficient top on the inside of vamp (including toes, instep, heel, carpus etc.) most
Small air line distance < 0.48cm.
It is preferred that, data acquisition device, including wearable device, camera, three-dimensional scanner;
It is preferred that, collect the port that device transmits data to Data Collection, can be mobile phone A PP etc radio port,
It can also be cable port;
It is preferred that, Data Collection collects device, including information pre-processing equipment, for screen effective information and exclude it is invalid
Information;
It is preferred that, Data Collection collects device, including DAF, for effective information to be sorted out, and to defect
Data are compensated;
It is preferred that, Data Collection collects device, including foot image file instant playback device, and the foot collected can be seen immediately
Portion's imaging data, gathered data can be supplemented at any time when finding data flaw;
It is preferred that, three-dimensional modeling apparatus, including at least level one data converting apparatus, data reforming unit can by pressure,
View data changes into the computer data for being available for editor;
It is preferred that, pattern recognition device and three-dimensional modeling apparatus can be integrated;
It is preferred that, three-dimensional modeling apparatus includes data storage device;
The three-dimensional modeling apparatus includes threedimensional model presentation device, it is preferred that threedimensional model presentation device has can be with
The function and corresponding construction at meaning adjustment visual angle, and including at least one display screen, can from any angle foot three-dimensional
Model;
It is preferred that, shoemaking device includes the equipment for making shoe tree;
It is preferred that, shoemaking device includes shoe body unit equipment, and it uses bonding, sews up, welds, riveting or above-mentioned side
Any combination mode of formula realizes the splicing of vamp and sole;
It is preferred that, shoemaking device includes making the equipment and sole-heel splicing apparatus of heel;
It is preferred that, vamp making apparatus and sole making apparatus can be integrated in same equipment, it is preferred that vamp makes
Equipment and sole making apparatus use 3D printing equipment;
It is preferred that, shoemaking device includes visual identity equipment, and it is used for the defect for recognizing inner sole;
It is preferred that, shoemaking device include inner sole corrective, its use cutting, and/or injection, and/or bonding and/
Or 3D printing method is used to correct inner sole defect, can be used for correction problem foot (as interior eight foots, outer eight foot, the rear strephexopodia,
Flat foot, talipes cavus, rear strephenopodia etc.) or for treat alleviate foot affliction (such as Plantar Fasciitis, nurse tumour, pain at metatarsal,
Neuralgia, diabetes etc.);
It is preferred that, shoemaking device includes inner sole grinding apparatus, and it is used to polishing smooth inner sole smooth.
The present invention relative to existing manufacturing method of shoes and the industrial process of system, the advantage is that, the present invention according to
The demand or foot feature at family, can at least while realizing following 9 kinds of key elements, (including user's request, product advantage, technology are excellent
Gesture) wherein 6 kinds:(1) shoes with sufficient type completely or approximate match;(2) foot health demand is met;(3) fabrication cycle is short;
(4) user's operation is time saving, convenient;(5) low manufacture cost;(6) shoes comfortableness is high;(7) shoes can be with orthopedic;(8) synthesis is examined
Consider foot image file, pressure distribution data, obtain complete foot data;(9) user individual (such as attractive in appearance) demand is met.
According to the present invention, when cordless collection foot image file data, way of contact collection human body are stood or walked
Plantar nervous arch data, wearable device collection foot data these three modes can be combined to obtain foot number
It is enough to be used in determining that foot is healthy, comfortable or orthopedic demand according to the foot data of, acquisition, when three kinds of modes are combined simultaneously
In use, the possibility for meeting the above-mentioned foot demand of user is greatly increased.
According to the present invention, the foot picture shot using camera can be most as little as 2, the home-confined i.e. available hand of user
Camera (such as mobile phone, camera) realization is held, whole process is only needed several seconds, enormously simplify operating time and the operation of user
Difficulty, the situation for the inconvenient directly contact user's foot that is particularly suitable for use in.
According to the present invention, the overall video of foot is shot using camera, the program is simple to operate, is particularly suitable for use in not
Convenient directly contact user's foot and the foot data of collection Duo Ren colonies.
According to the present invention, using electromagnetic wave, mechanical wave telemetry (including phase ranging method or pulse ranging method), pass through
It is long-range to obtain each point and collecting device different distance on sufficient face, then range data is changed into view data, when bright not enough
(being not enough to shooting clear image) and when being inconvenient to contact foot, the program has prominent practicality.Current existing shoemaking
Technology there is no such application.
According to the present invention, user can foot can with wearable device (including socks, deformable lapping, liquid or
The forms such as gas wrapping), it is accurate to obtain by judging the change in pattern on wearable device or sensor displacement thereon
The characteristics of foot data, program is that the higher foot data of precision, the threedimensional model set up accordingly and true foot can be achieved
Any local error be no more than 0.1mm.
According to the present invention, user can be using mobile phone A PP by foot data transfer to data processing platform (DPP), and a mobile phone can be same
Shi Shixian data acquisitions and data acquisition function, possess significant convenience.
According to the present invention, incomplete view data is fitted to primary foot threedimensional model, computerized algorithm pair is utilized
The defect of primary foot threedimensional model carries out image compensation reparation, and the program can utilize computer by image deflects instantaneously with smooth
Pattern is compensated, and related algorithm is integrated in computer systems, greatly improves modeling efficiency.
According to the present invention, produce it is completely the same or close to after consistent shoe tree with true foot size and shape, can be with
Traditional shoe-making process (such as footwear adhesion process, footwear hot vulcanization process, injection (penetrating) technique, mould pressing process, sewing tech) knot
The production process for completing shoes is closed, the docking property with existing shoemaking equipment, technology is strong, is conducive to quickly realizing industrialization.
According to the present invention, various kinds of sensors, communication device, charging device, temperature regulating device can be inserted in shoes in bottom, it is such
Device can realize wireless connection with intelligent control device (such as can be by mobile phone A PP), can in shoes integrated various functions (example
Exercise data, gait data, plantar pressure data, or regulation foot temperature are such as gathered, or realizes individual positioning etc.).
, can also according to the present invention it is possible to make inner sole using the material of different hardness, it would however also be possible to employ sandwich construction
Multiple boss are provided with above interior bottom, massage effect can be not only played accordingly and is comfortably experienced with lifting user, can also root
Foot local reduction or supercharging are realized according to the relevant medical knowledge such as sport dynamics, prevent the skills such as injury gained in sports, reduction athletic injury
Art effect.
According to the present invention, user is used after shoes, Consumer's Experience can be fed back to data processing platform (DPP), then according to user
Demand is adjusted to footwear construction.The technical scheme can realize the personalized comfort level requirement of user, also, can set up accordingly
Meet the Comfort Evaluation database of user's comfort requirement, for certain customers, data of feedback are that can be achieved to have for a long time
The comfort level experience of effect.
According to the present invention, the image deflects of foot threedimensional model require that maximum gauge is any no more than 0.05mm, inner sole
The maximum gauge of defect is no more than 0.5mm, there is no footwear art to propose such required precision at present.
According to the present invention, including foot threedimensional model is corrected, defect knowledge is carried out to inner sole using machine vision
Defect is not carried out not simultaneously the technical scheme such as to repair, there is no footwear art to include such precision controlling process at present.
According to the present invention, the interior bottom of shoes is fitted completely with true vola, any point and sufficient top on the inside of vamp (including foot
Toe, instep, heel, carpus etc.) minimum air line distance < 0.48cm, foot comfortableness provides significantly.Meanwhile, according to user
Demand, vamp and the minimum air line distance on sufficient top can suitably amplify and (such as may be adjusted to 0.5-1.5cm), can also be contracted to completely
Fitted with foot, shoes are as " long on pin ".
According to the present invention, image is restored using computer, strengthened, split, the combination energy of these three processing modes
It is enough effectively to make up the foot image file defect such as the aberration of image, brightness, leak, contrast, definition in source data, obtain using
In the outstanding view data of three-dimensional modeling.
Brief description of the drawings
Fig. 1 is a kind of customization footwear preparation method flow based on image recognition and dimensional Modeling Technology according to the present invention
Figure.
Fig. 2 is a kind of structure of customization footwear manufacturing system based on image recognition and dimensional Modeling Technology according to the present invention
Principle schematic.
Fig. 3 is the evaluation that Comfort Evaluation is carried out to " vamp inner side-sufficient top " minimum air line distance according to the present invention
Result figure.
Fig. 4 is a kind of mode of composition that view data is obtained according to the utilization filming apparatus of the present invention.
Fig. 5 is a kind of mode of composition that view data is obtained according to the utilization filming apparatus of the present invention.
Fig. 6 is a kind of shoe-pad produced according to 3D printing method therein of the invention.
Embodiment
1-2 is described further to embodiments of the invention below in conjunction with the accompanying drawings.
As shown in figure 1, the method for the making customization footwear based on image recognition and dimensional Modeling Technology of the present invention include with
Lower step:
(1) foot image file 100 is gathered:For obtaining the source datas such as foot image file data.Including seven kinds of modes, one is adopting
Foot image file data 101 are gathered with cordless, the second is foot image file data 102 are gathered using the way of contact, the third is
Using wearable device collection foot data 103, in addition four kinds of modes be using 101,102,103 3 kind of mode it is any
Close combination collection foot data.
(2) network collection foot data 200 are utilized:Including three kinds of modes, one is using cable network by foot data
It is collected into data processing platform (DPP) 201;The second is using wireless network by foot Data Collection to data processing platform (DPP) 202;Thirdly
It is to collect foot data using 201 and 202 combination.
(3) image recognition 300:Image is sequentially restored using computer, strengthens, split, changing into can
For the secondary data of three-dimensional modeling, specifically, i.e.,:
Image is carried out using computer to restore 301, the foot image file in source data is converted into digitized image, removed
Image noise, sets up the digital model of image degradation system, is then inversely derived by restored image;
Enhancing 302 is carried out to image using computer, cut using histogram equalization method or iconic model space reflection method
Foot image file and local detail in interference data in weak image, enhancing image;
Segmentation 303 is carried out to image using computer, by judging that the modes such as the discontinuous feature in image are carried on the back from shooting
Foot image file is extracted in scape, meanwhile, divide the image into form many sub-regions according to the gray scale in image, Texture eigenvalue, this
A little region constitutes the foot character shape region that ankle, heel, arch of foot, toes etc. are easy to significantly distinguish;
(4) three-dimensional modeling 400:Using secondary data carry out contours extract, model calculate, texture calculate, modeling rendering and
Render, form consistent with true foot or close to consistent foot threedimensional model, specifically, i.e.,:
Three-dimensional feature extracts mark 401, obvious characteristic point or profile in picture point cloud is found, then in other clouds
Search corresponding points simultaneously set up matching relationship, form the data basis of subsequent coordinates conversion and point cloud;
Coordinate Conversion and point cloud 402, are converted to D coordinates value, while will be many by the two-dimensional coordinate value of image data
Open foot image file and carry out a point cloud, form primary (may be with defective) foot threedimensional model;
Model global optimization 403, denoising, patch formation model defect (filling-up hole), segmentation office are carried out by primary foot measurements of the chest, waist and hips model
The global optimization processing such as portion's image, picture structure, stereo-picture surface be smooth, forms final foot threedimensional model.
(4) shoes 500 are made:Shoes are made according to foot threedimensional model, wherein:True foot is put on after shoes, shoes
Interior bottom fitted completely with true vola (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent).The process at least includes
Following several ways:Shoes 501 are directly printed using 3D printing technique;Make with true foot size and shape it is completely the same or
Close to consistent shoe tree, then made using footwear adhesion process, and/or footwear hot vulcanization process, and/or injection (penetrating) technique
Shoes, and/or mould pressing process make shoes, and/or sewing tech makes shoes 502;According to foot threedimensional model, without making
Shoe tree, the shape at each position of shoes is determined with visual determination (including naked eyes judge and computer vision judges), footwear are then utilized
Boots adhesion process, and/or footwear hot vulcanization process, and/or injection (penetrating) technique make shoes, and/or mould pressing process makes footwear
Son, and/or sewing tech make shoes 503;Shoes and shoe-pad are made respectively, shoe-pad is put into shoes final products 504 are made.
As shown in Fig. 2 the system of the making customization footwear of the present invention includes successively:
Data acquisition device I, its have image capture device 20, and/or foot pressure distributed constant collecting device 21,
And/or ripple (containing mechanical wave, electromagnetic wave) ranging imaging device 22, relevant parameter (pressure, the image in people vola 10 can be obtained
Deng), also with the port 23 for collecting the transmission data of device II to Data Collection;
Data Collection collects device II, and it has port 30, the data storage device that data are received from data acquisition device I
31, also with the port 32 that data are transmitted to pattern recognition device device III;
Pattern recognition device III, it has port 40, the image restoration dress for collecting the reception data of device II from Data Collection
Put 41, image intensifier device 42, image segmenting device 43;Also there is the port 44 that data are transmitted to three-dimensional modeling apparatus IV;
Three-dimensional modeling apparatus IV, it has port 50, the three-dimensional feature extraction dress that data are received from pattern recognition device III
51, coordinate transformation device 52, point cloud device 53, model global optimization device 54 are put, also with to shoemaking device V or V '
Transmit the port 55 of data;
Shoemaking device V and V ', V and V ' two kinds of Typical Representative for representing shoemaking device of the present invention, wherein:Ⅴ’
With port 60, the 3D printing equipment 61 that data are received from three-dimensional modeling apparatus;V has from three-dimensional modeling apparatus reception data
Port 70, high-precision vamp making apparatus 71, high-precision sole making apparatus 72.In the shoes that the shoemaking device is produced
Bottom is fitted (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent) completely with true vola, any point and foot on the inside of vamp
The minimum air line distance < 0.48cm on top (including toes, instep, heel, carpus etc.).V and V ' device produce into
Product shoes transfer to user 70 to wear.
It is preferred that, the system that footwear are customized according to the making of the present invention, its Data Collection, which collects device, can not include data
Storage device, i.e.,:Data acquisition device is joined directly together by internet with three-dimensional modeling apparatus.
It is preferred that, two foot image files are at least included in source data;
It is preferred that, collection foot image file includes any combination of one or more kinds of modes in the following manner:Using taking the photograph
Image device shoots the foot static images that each foot forms more than 3;Shooting foot formation using camera includes foot
The video of overall image;Complete foot 3-D view is gathered using spatial digitizer;Gathered using electromagnetic impulse telemetry
Complete foot 3-D view;Complete foot 3-D view is gathered using running safety method;Utilize mechanical wave impulse
Telemetry gathers complete foot 3-D view;Complete foot 3-D view is gathered using mechanical wave phase ranging method;
It is preferred that, it is that the foot image file that will be collected is transferred to number by special mobile phone A PP using network collection source data
According to processing platform;Or the network interconnection will be used between the collecting device of foot image file and terminal data processing platform, number is gathered every time
According to rear, collecting device it is automatic or under manual command by foot data transfer to the network terminal;
It is preferred that, the foot image file that source data is directly obtained using image capture device, and by plantar nervous arch
The vola view data that data are changed into;
It is preferred that, image is restored using computer, comprised the following steps:First, image conversion is carried out, by source number
Every foot image file in is transformed into the digitized image of space coordinate and the equal discretization of gray scale;Secondly, image is removed to make an uproar
Sound, that is, utilize probability density function statistics Gaussian noise, rayleigh noise, gamma noise, exponential distribution noise, uniformly distributed noise
After impulsive noise, pre-filtering is carried out to noise image and obtains noise variance, the high frequency characteristics further according to noise is filtered using average
Ripple method or median filtering method realize LPF;Again, image degradation system is set up, i.e., with mathematical function relationship g (x, y)=h
(x, y) × f (x, y)+n (x, y) expresses degradation model, and wherein x, y represents the two-dimensional coordinate point of digitized image, and g (x, y) is represented
Degraded image, h (x, y) represents degenrate function, and f (x, y) represents digitized static two dimensional image, and n (x, y) represents extraneous noise
Effect, then with digital modeling method confirmation degenrate function h (x, y) computational methods;Finally, according to h (x, y) result of calculation, make an uproar
The removal situation of sound variance and noise, is inversely derived to image degradation model, is finally completely eliminated or part abatement image
Space degenerate (fuzzy) and put degeneration (noise), obtain restored image;
It is preferred that, image is strengthened using computer, includes times of one or more kinds of modes in the following manner
Meaning combination:Gray Classification is carried out to foot image file, the gray level more than number of pixels spreaded using histogram equalization method, right
The few gray scale of number of pixels is compressed, and forms image local high-contrast;To foot image file carry out Gray Classification, then by each
Different grey-scale is transformed into different colours according to mapping function linearly or nonlinearly, makes the details of digitlization foot image file more
Plus substantially;Digitized image is subjected to Fourier transform, different colors are then transformed into by the mapping function of frequency, make numeral
The details for changing foot image file is more obvious;It is that parameter sets up foot HSV space mould with tone (H), saturation degree (S), lightness (V)
Type, the rgb space for digitizing foot image file is mapped on HSV space model, makes the details of digitlization foot image file brighter
It is aobvious;With cyan (C), carmetta (M), yellow (Y), black (K) are that parameter sets up foot CMYK spatial models, will digitize foot
The rgb space of image is mapped on CMYK spatial models, makes the details of digitlization foot image file more obvious;It is red with illumination (L)
Color is to the brightness range (a) of green, and the brightness range (b) of blueness to yellow sets up foot Lab space model for parameter, by numeral
The rgb space for changing foot image file is mapped on Lab space model, makes the details of digitlization foot image file more obvious;
It is preferred that, image is split using computer, both including being mutated according to gray scale, color change and texture mutation
Judge the discontinuity feature in image, and then extract the border of foot and shooting background in image, then foot image file from
Separated in shooting background, any combination of one or more kinds of processes in following process is included again:According to the ash of image
Degree divides the image into the subregion of non-overlapping copies, makes all subregion can be used in building threedimensional model;According to the shape of image
The subregion of non-overlapping copies is divided the image into, makes all subregion can be used in building threedimensional model;Will according to the texture of image
Image is divided into the subregion of non-overlapping copies, makes all subregion can be used in building threedimensional model;It will scheme according to the color of image
Subregion as being divided into non-overlapping copies, makes all subregion can be used in building threedimensional model;
It is preferred that, after the completion of image transformation step, image is compressed according to Joint Photographic Experts Group;
It is preferred that, digital modeling method confirms degenrate function h (x, y) computational methods, comprises the following steps:H (x, y) is entered
H (x, y) is obtained after row Fourier transform, expression formula H (x, y)=2.72^ (- ρ β (x2+y2) are utilized5/6) to H (x, y) assignment, then
Carry out obtaining h (x, y) last solution to the H (x, y) after assignment after reverse Fourier transform, wherein ρ represents digitized image office
Number of greyscale levels in portion's unit area, β represents the local picture element density of digitized image;
It is preferred that, using computer to image carry out it is enhanced during, to color, whether distortion is not specially required;
It is preferred that, histogram equalization method is to divide the image into some regions, with histogram equalization by region to image
Carry out computing and travel through all regions;Or some regions are divided the image into, each histogram equalization computing at least covers 2
Individual region;
It is preferred that, by using the Sobel operators or Roberts operators or granny rag Laplacian operater or Preveit of gray scale
The single order such as operator or Kirsch operators or Second Order Differential Operator detection discontinuity feature;
It is preferred that, collection foot image file includes any combination of one or more kinds of modes in the following manner:Foot is worn
On be equipped with the wearable device (such as socks) of sensor, foot image file is calculated by the deformation data for reading foot diverse location
Data;Foot is put on after the figuratum wearable device (such as socks) of print, and the pattern on wearable device is deformed upon, by clapping
Take the photograph the wearable device deformation pattern no less than 2 and obtain foot data;Foot is wrapped up with deformable material, it is deformable
The built-in sensor for being capable of mark position, foot image file data are calculated by reading sensing station and displacement in material;Will
Foot is put into liquid or gas, the sensor of mark position is capable of built in liquid, gas, by reading sensing station and position
Shifting calculates foot image file data;
It is preferred that, three-dimensional feature extracts mark, is to find the obvious characteristic point or profile in picture point cloud, then at other
Corresponding points are searched in point cloud and matching relationship is set up, and it is extracted or mark mode includes the one or more side in the following manner
Any combination of formula:Extract or mark foot image file in sufficient edge, toes, heel, ankle, arch of foot etc. have notable gray scale,
The picture point of chromatic characteristic, corresponding points are searched in other clouds and matching relationship is set up;Extract curvature, the method in foot curved surface
The invariant features such as vector, the normal vector angle of point-to-point transmission, corresponding points are searched in other clouds and matching relationship is set up;Mark foot
1 points of distance relation two-by-two in portion's image, corresponding points are searched in other clouds and matching relationship is set up;Treat spelling
The two panels point cloud or wherein a piece of cloud connect is sampled, and it is right on two panels point cloud then to be made using icp algorithms by iterative calculation
The mean square error that should put to or correspond to point-to-plane distance is minimum, and then determines matching relationship;With the curvature of curved surface be characterized a little, with
Angle, distance are constraints, and point cloud matching relation is set up using 3 rotary process between a cloud;
It is preferred that, Coordinate Conversion and point cloud, are that the two-dimensional coordinate value of image data is converted into D coordinates value, and
Multiple images are matched using characteristic point or profile as binding site, the step of forming preliminary three-dimensional model, its Coordinate Conversion is pressed
Sequencing includes:Transformation of scale, that is, adjusting the proportionate relationship of each image is consistent the engineer's scale of all images;Two dimension is sat
Mark is converted into three-dimensional coordinate;
It is preferred that, include any combination of one or more modes in following optimal way:Denoising, patch formation model defect
(filling-up hole), segmentation topography, picture structure, stereo-picture surface are smooth;
It is preferred that, curvature is Gaussian curvature or average curvature;
It is preferred that, three-dimensional feature point is extracted or mark is, it is necessary to reject bad point matching;
It is preferred that, two panels point cloud to be spliced or wherein a piece of cloud are sampled, including uniform sampling method, adopts at random
Sample method, normal vector sampling method;
It is preferred that, after iterative calculation is finished, unreliable or incompatible correspondence is removed using the way of restraint for icp algorithms
Point pair or point face, its way of restraint is a kind of the way of restraint based on curved surface features, that is, constrains curvature, normal direction in foot curved surface
The invariant features such as amount, the normal vector angle of point-to-point transmission;
It is preferred that, using Solidworks, Pro/E, UG, CAD, Adobe Fuse CC, 3Ds Max, SketchUp,
Other 3 d modeling softwares of Revit or any realize three-dimensional modeling;
It is preferred that, inserted in the interior bottom of shoes pressure sensor, and/or acceleration transducer, and/or positioner and/
Or alarming device, and/or wireless communication apparatus, and/or power supply and charging device, and/or temperature regulating device;
It is preferred that, the interior bottom of shoes is detachable apparatus, it is preferable that the interior bottom of shoes is shoe-pad, it is preferable that below shoe-pad
If headspace, pad can be inserted in headspace according to the difference in height of human body double lower limb, human body is got on one's shoes can after son
Ensure balance, can be used for human body and increase;
It is preferred that, (i.e. footwear of being fitted completely with true foot top (including toes, instep, heel, carpus etc.) on the inside of vamp
The inner side of son is corresponding consistent with the shape on true sufficient top);
It is preferred that, fixed additional materials are set up above the interior bottom of shoes, make the interior bottom and true vola of shoes incomplete
Laminating, for correcting problem foot (eight foots, outer eight foot, the rear strephexopodia, flat foot, talipes cavus, rear strephenopodia in such as) or being used for
Foot affliction (such as Plantar Fasciitis, nurse tumour, pain, neuralgia, diabetes at metatarsal) is alleviated in treatment;
It is preferred that, 0.5cm projection is no more than above the interior bottom of shoes provided with multiple height, massage effect can be played,
Preferably, additional materials are corresponding with the acupuncture point of sole portion or whole;
It is preferred that, user is used after shoes, Consumer's Experience is fed back into data processing platform (DPP), then according to user's request pair
Footwear construction is adjusted;
It is preferred that, multiple closed individual cavities are provided with the interior bottom of shoes, shoes quality can be mitigated, can also be in human body
Increase cushioning effect during strenuous exercise;
It is preferred that, the interior bottom of shoes is made using wear-resisting, waterproof, with ductility and elastic material, is preferably true
Skin, rubber, polyurethanes (PU), vinyl acetate copolymer (EVA), MD, TPR, TPUE (TPU),
Polyvinyl chloride (PVC), TR, imitation leather bottom, BPU, thermoplastic elastic rubber-sole, makrolon (PC) resin, acrylonitrile-butadiene-benzene
Ethylene copolymer (ABS), butadiene-styrene rubber (SBR), and their arbitrary composition;
It is preferred that, the interior bottom of shoes can be sandwich construction, it is preferable that each layer in interior bottom of shoes uses different material systems
Into;
It is preferred that, height heels are no more than 4cm, preferably 1cm;
It is preferred that, foot threedimensional model is corrected, when foot threedimensional model is amplified to true foot size with
Untill true foot is no more than 0.5mm compared to the image deflects maximum gauge at any place, preferably 0.1mm;
It is preferred that, defect recognition is carried out to inner sole using machine vision, when defects with diameters is more than 1mm in footwear, used
Cutting, and/or injection, and/or bonding, and/or 3D printing method amendment sole defect;
It is preferred that, three-dimensional modeling uses CCD optical modeling techniques;
It is preferred that, when obtaining foot image file using filming apparatus, also include foot office using the foot data of network collection
Portion or the length information of whole;
It is preferred that, sensor of the contact surface provided with multiple detectable deformation or pressure change of shoes and foot can detect
Go out gait change or foot hyperblastosis;
It is preferred that, the tensile strength at bottom is not less than 8.5kgf/mm in shoes2, it is preferred that the tensile strength at bottom is not in shoes
Less than 10kgf/mm2;
It is preferred that, the bending modulus at bottom is not less than 140kgf/mm in shoes2, it is preferred that the bending modulus at bottom is not in shoes
Less than 180kgf/mm2;
It is preferred that, the tensile strength at bottom is not less than 8.5kgf/mm in shoes2, it is preferred that the tensile strength at bottom is not in shoes
Less than 10kgf/mm2;
It is preferred that, the bending modulus at bottom is not less than 140kgf/mm in shoes2, it is preferred that the bending modulus at bottom is not in shoes
Less than 180kgf/mm2;
It is preferred that, data acquisition device, including wearable device, camera, three-dimensional scanner;
It is preferred that, collect the port that device transmits data to Data Collection, can be mobile phone A PP etc radio port,
It can also be cable port;
It is preferred that, Data Collection collects device, including information pre-processing equipment, for screen effective information and exclude it is invalid
Information;
It is preferred that, Data Collection collects device, including DAF, for effective information to be sorted out, and to defect
Data are compensated;
It is preferred that, Data Collection collects device, including foot image file instant playback device, and the foot collected can be seen immediately
Portion's imaging data, gathered data can be supplemented at any time when finding data flaw;
It is preferred that, three-dimensional modeling apparatus, including at least level one data converting apparatus, data reforming unit can by pressure,
View data changes into the computer data for being available for editor;
It is preferred that, pattern recognition device and three-dimensional modeling apparatus can be integrated;
It is preferred that, three-dimensional modeling apparatus includes data storage device;
The three-dimensional modeling apparatus includes threedimensional model presentation device, it is preferred that threedimensional model presentation device has can be with
The function and corresponding construction at meaning adjustment visual angle, and including at least one display screen, can from any angle foot three-dimensional
Model;
It is preferred that, shoemaking device includes the equipment for making shoe tree;
It is preferred that, shoemaking device includes shoe body unit equipment, and it uses bonding, sews up, welds, riveting or above-mentioned side
Any combination mode of formula realizes the splicing of vamp and sole;
It is preferred that, shoemaking device includes making the equipment and sole-heel splicing apparatus of heel;
It is preferred that, vamp making apparatus and sole making apparatus can be integrated in same equipment, it is preferred that vamp makes
Equipment and sole making apparatus use 3D printing equipment;
It is preferred that, shoemaking device includes visual identity equipment, and it is used for the defect for recognizing inner sole;
It is preferred that, shoemaking device include inner sole corrective, its use cutting, and/or injection, and/or bonding and/
Or 3D printing method is used to correct inner sole defect, can be used for correction problem foot (as interior eight foots, outer eight foot, the rear strephexopodia,
Flat foot, talipes cavus, rear strephenopodia etc.) or for treat alleviate foot affliction (such as Plantar Fasciitis, nurse tumour, pain at metatarsal,
Neuralgia, diabetes etc.);
It is preferred that, shoemaking device includes inner sole grinding apparatus, and it is used to polishing smooth inner sole smooth.
Embodiment 1:
Adopt and make customization footwear with the following method:Foot image file is gathered, source data is obtained;Utilize network collection source data;Figure
As identification, i.e., image sequentially restored using computer, strengthened, split, changed into and be available for the two of three-dimensional modeling
Secondary data;Three-dimensional modeling, i.e., carry out three-dimensional feature extraction mark, Coordinate Conversion and point cloud, model complete using secondary data
Office's optimization, forms consistent with true foot or close to consistent foot threedimensional model;Made according to foot threedimensional model
Shoes, the interior bottoms of shoes is fitted (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent), vamp completely with true vola
The minimum air line distance < 0.48cm on inner side any point and sufficient top (including toes, instep, carpus etc.).
Particularly, the present invention utilizes statistics, it is determined that in vamp by the investigation to 100 user experience datas
When the minimum air line distance of side and sufficient top is improved to " 0.48cm " (average data), users'comfort is remarkably decreased, relevant to adjust
Look into process and data are as follows:
Investigation method:At random (factor such as age, sex, height, body weight is random) choose 100 people, to it is each per capita
10 pairs of shoes of respective production.Shoes feature includes:(1) sole of every shoes is matched completely with corresponding vola shape;(2) it is every double
" vamp inner side-sufficient top " minimum air line distance (L) of shoes is consistent;The L of (3) 10 pairs of shoes is from 0.1cm-1.0cm structures
Into tolerance be 0.1cm arithmetic progression, i.e.,:The L of 1st pair, the 2nd pair ... the 10th pair of footwear be respectively 0.1cm, 0.2cm,
0.3cm……1.0cm.Everyone corresponding 10 pairs of footwear, which upset order, makes everyone can not know the L of each pair of shoes, then often
Individual tries 10 pairs of footwear corresponding with oneself on respectively, tries walking distance on every time and is not less than 1000 meters, and to every pair of shoes with " no
It is comfortable " whether (U), " uncertain comfortable " (N), the shoes of " comfortable " (W) be marked.Evaluation result is as shown in figure 3, wherein:
Abscissa represents L, and ordinate represents the evaluation quantity that different L is obtained.According to Fig. 3, W to U notable interface point is in 0.39cm
Left and right, U to N notable interface point is in 0.48cm or so.Accordingly, and consider random individual impression uncertainty, sentence
It is disconnected as follows:During L=0.48cm, the individual impression of shoes user is tended to depart from " comfortable " state, the feasibility of the technical scheme
With statistics evidence.From another perspective, when L is fluctuated near 0.48cm data, the individual impression hair of shoes user
Raw unexpected transformation, the i.e. shoemaking scheme of L=0.48cm can use to obtain unexpected technique effect.
Embodiment 2:
Adopt and make customization footwear with the following method:User claps to obtain 4 foot pictures using photographing function of mobile phone, and user surveys
Amount obtains foot length information;Utilize 4 foot pictures of network collection and foot length information;Image is sequentially entered using computer
Row restores, strengthened, segmentation, changes into the secondary data for being available for three-dimensional modeling;Three-dimensional feature, which is carried out, using secondary data extracts mark
Note, Coordinate Conversion and point cloud, model global optimization, form consistent with true foot or close to consistent foot three
Dimension module;Shoes are made according to foot threedimensional model, the interior bottoms of shoes is fitted (the i.e. interior bottom of shoes and very completely with true vola
The shape correspondence at full bottom is consistent), the minimum straight line of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.)
Apart from < 0.48cm.
In the present embodiment, user clap the modes of composition of 4 foot pictures has a variety of, to obtain complete foot image file number
According to.Preferably meet following basic demand:(1) whole plantars surfaces of toe (sole) can be included;(2) the instep face of instep most convex line both sides,
Respectively need at least one photo;(3) combination of all photos, can at least include the 95% of all its effective areas of sufficient face.
Fig. 4 illustrates a kind of implementation for shooting 4 foot pictures, and wherein photo 1 and photo 4 contains instep respectively
The instep face of most convex line both sides, photo 2 and the combination of photo 3 can be comprising whole plantars surfaces of toe, and four photo combinations can be comprising complete
More than the 95% of its effective area of portion foot face.
Fig. 5 illustrates another implementation for shooting 4 foot pictures, and wherein photo 6, photo 7 and photo 8 combines energy
The instep face of instep most convex line both sides is enough included, photo 5 and the combination of photo 6 can be combined comprising whole plantars surfaces of toe, four photos
More than the 95% of all its effective areas of sufficient face can be included.
Shooting style described by the present embodiment, sufficient face image (such as contact of adjacent toe to can not remotely obtain
Face) information do not specially require.
Embodiment 3:
Adopt and make customization footwear with the following method:User claps to obtain 2 foot pictures using photographing function of mobile phone, and user surveys
Amount obtains foot length information;Utilize 2 foot pictures of network collection and foot length information;Image is sequentially entered using computer
Row restores, strengthened, segmentation, changes into the secondary data for being available for three-dimensional modeling;Three-dimensional feature, which is carried out, using secondary data extracts mark
Note, Coordinate Conversion and point cloud, model global optimization, form consistent with true foot or close to consistent foot three
Dimension module;Shoes are made according to foot threedimensional model, the interior bottoms of shoes is fitted (the i.e. interior bottom of shoes and very completely with true vola
The shape correspondence at full bottom is consistent), the minimum straight line of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.)
Apart from < 0.48cm.
In the present embodiment, the foot information that its 2 foot picture clapped of untrained user are recorded is typically endless
It is whole, therefore be necessary to supplement defect image during three-dimensional modeling, repair.To the greatest extent may be used to be obtained using two photos
Can completely foot information, typically has claimed below to user's mode of composition:(1) clapped respectively from instep most convex line both sides and take one
Photo;(2) wherein one photo had not only included the image of whole plantars surfaces of toe (sole) but also had included the image of instep most convex line side;
(3) wherein another photo includes all images of instep most convex line opposite side.
Photo 1 and photo 3 in Fig. 4 are combined, and can meet above-mentioned photographing request;In addition, photo 5 and the phase of photo 7 in Fig. 5
With reference to above-mentioned photographing request can also be met.
Embodiment 4:
Adopt and make customization footwear with the following method:User claps to obtain 3 foot pictures using photographing function of mobile phone, and user surveys
Amount obtains foot length information;Utilize 3 foot pictures of network collection and foot length information;Image is sequentially entered using computer
Row restores, strengthened, segmentation, changes into the secondary data for being available for three-dimensional modeling;Three-dimensional feature, which is carried out, using secondary data extracts mark
Note, Coordinate Conversion and point cloud, model global optimization, form consistent with true foot or close to consistent foot three
Dimension module;Shoes are made according to foot threedimensional model, the interior bottoms of shoes is fitted (the i.e. interior bottom of shoes and very completely with true vola
The shape correspondence at full bottom is consistent), the minimum straight line of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.)
Apart from < 0.48cm.
In the present embodiment, user clap the modes of composition of 3 foot pictures has a variety of, to obtain complete foot image file number
According to.Preferably meet following basic demand:(1) whole plantars surfaces of toe (sole) can be included;(2) the instep face of instep most convex line both sides,
Respectively need at least one photo.
Photo 1, photo 3 and photo 4 in Fig. 4 are combined, and can meet above-mentioned photographing request.
Embodiment 5:
Adopt and make customization footwear with the following method:Foot image file is gathered, source data is obtained;Utilize network collection source data;Figure
As identification, i.e., image sequentially restored using computer, strengthened, split, changed into and be available for the two of three-dimensional modeling
Secondary data;Three-dimensional modeling, i.e., carry out three-dimensional feature extraction mark, Coordinate Conversion and point cloud, model complete using secondary data
Office's optimization, forms consistent with true foot or close to consistent foot threedimensional model;Made according to foot threedimensional model
The shoe-pad fitted completely with true vola, the minimum of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.) is straight
Linear distance < 0.48cm, are put into shoes by shoe-pad and final products are made.
Fig. 6 illustrates one kind and utilizes shoe-pad produced by the invention, and photo 600 is shoe-pad and the contact surface view in vola, is somebody's turn to do
Contact surface is fitted completely with true vola, and photo 601 is shoe-pad bottom view, and photo 602 and photo 603 are two side views of shoe-pad
Figure.The shoe-pad is provided with relatively low guardrail on the outside of sufficient side provided with higher guardrail, on the inside of sufficient side, plays a part of including
Prevent from not influenceing action while athletic injury (such as twisting leg).
Embodiment 6:
Adopt and make customization footwear with the following method:User claps to obtain 6 foot pictures using photographing function of mobile phone, and picture includes
Object of reference knowable to the length being placed in beside foot;The foot data collected are transferred to number by user by special mobile phone A PP
According to processing platform;View data is converted into primary foot threedimensional model using UG modeling softwares, to primary foot threedimensional model
Carry out image compensation, repair, obtain final foot threedimensional model, foot threedimensional model is consistent with true foot;According to
Foot threedimensional model makes shoe tree, and shoes are made further according to shoe tree, and the interior bottom of shoes finished shoe is fitted completely with true vola,
The minimum air line distance on any point and sufficient top is 0.3cm on the inside of vamp;It is no more than above the interior bottom of shoes provided with multiple height
0.2cm projection, it is each raised corresponding with the acupuncture point in vola;Pressure sensor is inserted in the interior bottom of shoes;Height heels are
2cm;It is provided with the interior bottom of shoes in multiple closed separate chambers, chamber and liquid is housed, comfort level can be increased, can also be in people
The tensile strength for increasing bottom in cushioning effect, shoes during body strenuous exercise is 12kgf/mm2。
Embodiment 7:
Adopt and make customization footwear with the following method:Foot image file is gathered, source data is obtained;Utilize network collection source data;Figure
As identification, the image being collected into is changed, every foot image file in source data is transformed into space coordinate and gray scale
The digitized image of equal discretization;After the completion of image transformation step, image is compressed according to Joint Photographic Experts Group;Image is removed to make an uproar
Sound, that is, utilize probability density function statistics Gaussian noise, rayleigh noise, gamma noise, exponential distribution noise, uniformly distributed noise
After impulsive noise, pre-filtering is carried out to noise image and obtains noise variance, the high frequency characteristics further according to noise is filtered using average
Ripple method or median filtering method realize LPF;Again, image degradation system is set up, i.e., with mathematical function relationship g (x, y)=h
(x, y) × f (x, y)+n (x, y) expresses degradation model, and wherein x, y represents the two-dimensional coordinate point of digitized image, and g (x, y) is represented
Degraded image, h (x, y) represents degenrate function, and f (x, y) represents digitized static two dimensional image, and n (x, y) represents extraneous noise
Effect.H (x, y) is obtained after carrying out Fourier transform to h (x, y), expression formula H (x, y)=2.72^ (- ρ β (x2+y2) are utilized5/6)
To H (x, y) assignment, then carry out obtaining h (x, y) last solution, wherein ρ generations to the H (x, y) after assignment after reverse Fourier transform
Number of greyscale levels in table digitized image local unit area, β represents the local picture element density of digitized image;Then, according to h
The removal situation of the result of calculation of (x, y), noise variance and noise, is inversely derived to image degradation model, it is final completely
Eliminate or part cuts down the space degeneration (fuzzy) of image and point is degenerated (noise), obtain restored image.Recycle computer pair
Image is strengthened, and is carried out Gray Classification to foot image file, some regions is divided the image into using histogram equalization method, with straight
Side's figure equalization carries out computing to image by region and travels through all regions, the gray level more than number of pixels is spreaded,
The gray scale few to number of pixels is compressed, and image local high-contrast is formed, by each different grey-scale according to linear or non-
Linear mapping function is transformed into different colours, makes the details of digitlization foot image file more obvious;Then, by digital map
As carrying out Fourier transform, different colors are then transformed into by the mapping function of frequency, make the details of digitlization foot image file
It is more obvious;With illumination (L), the red brightness range (a) to green, the brightness range (b) of blueness to yellow is set up for parameter
Foot Lab space model, the rgb space for digitizing foot image file is mapped on Lab space model, makes digitlization foot image file
Details it is more obvious.Recycle computer image is split, using the Sobel operators or Roberts operators of gray scale,
Or in granny rag Laplacian operater or the single order such as Preveit operators or Kirsch operators or Second Order Differential Operator progress detection image
Discontinuity feature, and then the border of foot and shooting background in image is extracted, then foot image file is divided from shooting background
Separate out and, and the subregion of non-overlapping copies is divided the image into according to the gray scale of image, shape, texture, color, make all subregion
It can be used in building threedimensional model.Two panels point cloud to be spliced or wherein a piece of cloud are sampled again, then calculated using icp
Method by iterative calculation make corresponding points on two panels point cloud to or correspondence point-to-plane distance mean square error it is minimum, iterative calculation is finished
Afterwards, the invariant features such as curvature, normal vector, normal vector angle of point-to-point transmission in constraint foot curved surface, to remove it is unreliable or
Incompatible corresponding points pair or point face, and then determine matching relationship;It is characterized a little with the curvature of curved surface simultaneously, with angle, distance
For constraints, point cloud matching relation is set up using 3 rotary process between a cloud;Extract curvature, the normal direction in foot curved surface
The invariant features such as amount, the normal vector angle of point-to-point transmission, corresponding points are searched in other clouds and matching relationship is set up;Mark simultaneously
1 points of distance relation two-by-two in foot image file, corresponding points are searched in other clouds and matching relationship is set up.Reject
After bad point matching, adjusting the proportionate relationship of each image is consistent the engineer's scale of all images, then by view data two dimension
Coordinate Conversion is into three-dimensional coordinate.Then denoising, patch formation model defect (filling-up hole), segmentation topography, image knot are carried out to image
The global optimization processing such as structure, stereo-picture surface be smooth, obtains consistent with true foot or close to consistent foot
Threedimensional model;Shoes are made according to foot threedimensional model, the interior bottoms of shoes fitted completely with true vola (i.e. the interior bottom of shoes and
The shape correspondence in true vola is consistent), the minimum of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.) is straight
Linear distance < 0.48cm.
Embodiment 8:
Adopt and make customization footwear with the following method:Complete foot 3-D view is gathered using electromagnetic impulse telemetry;With
The foot data collected are passed through cable-network transmission to data processing platform (DPP) by family;Utilize Adobe Fuse CC modeling softwares
View data is converted into primary foot threedimensional model, image compensation is carried out to primary foot threedimensional model, is repaired, obtains final
Foot threedimensional model, foot threedimensional model is consistent with true foot;Make and true foot size and shape complete one
Cause or close to consistent shoe tree, then make shoes using footwear adhesion process;The interior bottom of shoes is fitted completely with true vola,
The minimum air line distance on any point and sufficient top is 0.2cm on the inside of vamp;Positioner is inserted in interior bottom, height heels are 3cm;
Multiple closed individual cavities are provided with the interior bottom of shoes, shoes quality can be mitigated, can also be increased in human body strenuous exercise
Cushioning effect;The bending modulus at bottom is 190kgf/mm in shoes2, user use shoes after, Consumer's Experience is fed back at data
Platform, is then adjusted according to user's request to footwear construction.
Embodiment 9:
Adopt and make customization footwear with the following method:Complete foot 3-D view is gathered using running safety method;With
The foot data collected are passed through wireless network transmissions to data processing platform (DPP) by family;Using 3Ds Max modeling softwares by image
Data conversion carries out image compensation to primary foot threedimensional model, repaired, obtain final foot into primary foot threedimensional model
Threedimensional model, foot threedimensional model is consistent with true foot;Make completely the same or connect with true foot size and shape
Nearly consistent shoe tree, then makes shoes using footwear hot vulcanization process;The interior bottom of shoes is fitted completely with true vola, vamp
Inner side is appointed fits completely with sufficient top;Alarming device is inserted in interior bottom;The interior bottom of shoes can be three-decker, and it is adopted at the middle and upper levels
Made of polyurethanes (PU), lower floor is made using vinyl acetate copolymer (EVA);
Embodiment 10:
Adopt and make customization footwear with the following method:Complete foot 3-D view is gathered using mechanical wave pulse ranging method;With
The foot data collected are passed through wireless network transmissions to data processing platform (DPP) by family;Plantar pressure is divided using computerized algorithm
Cloth data change into view data, and view data is converted into primary foot threedimensional model using SketchUp modeling softwares, right
Primary foot threedimensional model carries out image compensation, repaired, then foot threedimensional model is corrected, until foot threedimensional model is put
Untill the image deflects maximum gauge at any place is no more than 0.1mm compared with true foot greatly when true foot size;Make
It is completely the same or close to consistent shoe tree with true foot size and shape, then make shoes using injection (penetrating) technique;Footwear
The interior bottom of son is fitted completely with true vola, and the minimum air line distance on any point and sufficient top is 0.1cm on the inside of vamp;In interior bottom
Insert wireless communication apparatus and power supply and charging device;Inner sole is polished, makes inner sole smooth.
Embodiment 11:
Adopt and make customization footwear with the following method:Complete foot 3-D view is gathered using mechanical wave phase ranging method;With
The foot data collected are passed through wireless network transmissions to data processing platform (DPP) by family;Using Revit modeling softwares by picture number
According to primary foot threedimensional model is converted into, image compensation is carried out to primary foot threedimensional model, is repaired, final foot three is obtained
Dimension module, foot threedimensional model is consistent with true foot;Directly print shoe-pad using 3D printing technique, the interior bottom of shoes with
Fit completely in true vola;Weave vamp using numerical control weaving, on the inside of vamp the minimum straight line on any point and sufficient top away from
From for 0.05cm;The tensile strength that bottom in temperature control device, shoes is inserted in interior bottom is 11kgf/mm2。
Embodiment 12:
Adopt and make customization footwear with the following method:User obtains foot picture using photographing function of mobile phone;User will collect
Foot data pass through wireless network transmissions to data processing platform (DPP);Foot threedimensional model is built using CCD optical modeling techniques,
Foot threedimensional model is consistent with true foot;Make completely the same or close to consistent footwear with true foot size and shape
Last carving, then makes shoes using mould pressing process, wherein, the making precision of inner sole is controlled using laser cutting mode, is made in footwear
The maximum gauge of any defect in bottom is no more than 1mm;Defect recognition is carried out to inner sole using machine vision, when defects with diameters in footwear
During more than 1mm, using 3D printing method amendment inner sole defect, the interior bottom of shoes is set to be fitted completely with true vola, in vamp
Fitted completely with sufficient top side;The interior bottom of shoes is detachable apparatus.
Embodiment 13:
Adopt and make customization footwear with the following method:Foot image file is gathered, source data is obtained;Utilize network collection source data;Figure
As identification, the image being collected into is changed, every foot image file in source data is transformed into space coordinate and gray scale
The digitized image of equal discretization;After the completion of image transformation step, image is compressed according to Joint Photographic Experts Group;Image is removed to make an uproar
Sound, that is, utilize probability density function statistics Gaussian noise, rayleigh noise, gamma noise, exponential distribution noise, uniformly distributed noise
After impulsive noise, pre-filtering is carried out to noise image and obtains noise variance, the high frequency characteristics further according to noise is filtered using average
Ripple method or median filtering method realize LPF;Again, image degradation system is set up, i.e., with mathematical function relationship g (x, y)=h
(x, y) × f (x, y)+n (x, y) expresses degradation model, and wherein x, y represents the two-dimensional coordinate point of digitized image, and g (x, y) is represented
Degraded image, h (x, y) represents degenrate function, and f (x, y) represents digitized static two dimensional image, and n (x, y) represents extraneous noise
Effect.H (x, y) is obtained after carrying out Fourier transform to h (x, y), expression formula H (x, y)=2.72^ (- ρ β (x2+y2) are utilized5/6)
To H (x, y) assignment, then carry out obtaining h (x, y) last solution, wherein ρ generations to the H (x, y) after assignment after reverse Fourier transform
Number of greyscale levels in table digitized image local unit area, β represents the local picture element density of digitized image;Then, according to h
The removal situation of the result of calculation of (x, y), noise variance and noise, is inversely derived to image degradation model, it is final completely
Eliminate or part cuts down the space degeneration (fuzzy) of image and point is degenerated (noise), obtain restored image.Recycle computer pair
Image is strengthened, and is carried out Gray Classification to foot image file, some regions is divided the image into using histogram equalization method, with straight
Side's figure equalization carries out computing to image by region and travels through all regions, the gray level more than number of pixels is spreaded,
The gray scale few to number of pixels is compressed, and image local high-contrast is formed, by each different grey-scale according to linear or non-
Linear mapping function is transformed into different colours, makes the details of digitlization foot image file more obvious;Then, by digital map
As carrying out Fourier transform, different colors are then transformed into by the mapping function of frequency, make the details of digitlization foot image file
It is more obvious;With cyan (C), carmetta (M), yellow (Y), black (K) are that parameter sets up foot CMYK spatial models, by numeral
The rgb space for changing foot image file is mapped on CMYK spatial models, makes the details of digitlization foot image file more obvious.Recycle
Computer is split to image, using gray scale Sobel operators or Roberts operators or granny rag Laplacian operater or
The single order such as Preveit operators or Kirsch operators or Second Order Differential Operator carry out the discontinuity feature in detection image, and then
The border of foot and shooting background in image is extracted, then foot image file is separated from shooting background, and according to image
Gray scale, shape, texture, color divide the image into the subregions of non-overlapping copies, make all subregion can be used in building it is three-dimensional
Model.The invariant features such as curvature, normal vector, the normal vector angle of point-to-point transmission in foot curved surface are extracted again, in other clouds
Search corresponding points simultaneously set up matching relationship;While 1 points in foot image file of distance relation two-by-two is marked, in other points
Corresponding points are searched in cloud and matching relationship is set up.Reject after bad point matching, adjusting the proportionate relationship of each image makes all images
Engineer's scale is consistent, and view data two-dimensional coordinate then is converted into three-dimensional coordinate.Then denoising, repairing mould are carried out to image
The global optimization processing such as type defect (filling-up hole), segmentation topography, picture structure, stereo-picture surface be smooth, obtain with very
Real foot is consistent or close to consistent foot threedimensional model;Shoes, the interior bottom of shoes are made according to foot threedimensional model
Fitted completely with true vola (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent), on the inside of vamp on any point and foot
The minimum air line distance < 0.48cm in portion (including toes, instep, carpus etc.).
Embodiment 14:
Adopt and make customization footwear with the following method:The sensor that user's foot is put on the socks for being equipped with sensor, socks will
The deformation data of foot diverse location is transferred to data processing platform (DPP);Deformation data is changed into picture number using computerized algorithm
According to using CCD optical modeling techniques structure foot threedimensional model, foot threedimensional model is consistent with true foot;Make with
True foot size and shape is completely the same or close to consistent shoe tree, then using mould pressing process making shoes, shoes it is interior
Fitted completely with true vola at bottom.
Embodiment 15:
Adopt and make customization footwear with the following method:User's foot is put on after the figuratum wearable device (such as socks) of print, can
Pattern on object wearing device is deformed upon, and foot number is obtained by the wearable device deformation pattern shot no less than 2
According to;Deformation data is changed into view data using computerized algorithm, the three-dimensional mould of foot is built using CCD optical modeling techniques
Type, foot threedimensional model is consistent with true foot;Make completely the same or close consistent with true foot size and shape
Shoe tree, then using mould pressing process make shoes, the interior bottom of shoes is fitted completely with true vola.
Embodiment 16:
Adopt and make customization footwear with the following method:User's foot is put on after the figuratum wearable device (such as socks) of print, can
Pattern on object wearing device is deformed upon, and foot number is obtained by the wearable device deformation pattern shot no less than 2
According to;Deformation data is changed into view data using computerized algorithm, the three-dimensional mould of foot is built using CCD optical modeling techniques
Type, foot threedimensional model is consistent with true foot;Make completely the same or close consistent with true foot size and shape
Shoe tree, then make shoes using sewing tech, the interior bottoms of shoes is shoe-pad;Shoe-pad is fitted completely with true vola;Shoe-pad
Lower section sets headspace, and pad is inserted in headspace according to the difference in height of human body double lower limb, human body is got on one's shoes energy after son
It is enough to ensure balance.
Embodiment 17:
Adopt and make customization footwear with the following method:User wraps up built-in energy in foot, deformable material with deformable material
The sensor of enough mark positions, foot data are calculated by reading sensing station and displacement;Utilize computerized algorithm
Deformation data is changed into view data, foot threedimensional model is built using CCD optical modeling techniques, foot threedimensional model with it is true
Real foot is consistent;Make completely the same or close to consistent shoe tree with true foot size and shape, then using being molded
Technique makes shoes, and the interior bottom of shoes is shoe-pad;Shoe-pad is fitted completely with true vola;The contact surface of shoe-pad and foot is provided with more
Individual detectable deformation or the sensor of pressure change, can detect that gait change or foot hyperblastosis.
Embodiment 18:
Adopt and make customization footwear with the following method:User puts foot in liquid into, largely being capable of mark position built in liquid
Sensor of vacillating, sensor is uniformly distributed in a liquid, and foot is calculated by reading sensing station and displacement
Data;Deformation data is changed into view data using computerized algorithm, the three-dimensional mould of foot is built using CCD optical modeling techniques
Type, foot threedimensional model is consistent with true foot;Make completely the same or close consistent with true foot size and shape
Shoe tree, then make shoes using mould pressing process, the interior bottoms of shoes is shoe-pad;Shoe-pad is fitted completely with true vola;Shoe-pad
The footwear bottom being in contact with it is made of the material of different-stiffness.
Embodiment 19:
Adopt and make customization footwear with the following method:The sensor that user's foot is put on the socks for being equipped with sensor, socks will
The deformation data of foot diverse location is transferred to data processing platform (DPP);Deformation data is changed into picture number using computerized algorithm
According to using CCD optical modeling techniques structure foot threedimensional model, foot threedimensional model is consistent with true foot;Do not make
Shoe tree, and use machine vision to judge that combining automatic carving device makes sole, the interior bottom of shoes is fitted completely with true vola;
The minimum air line distance that any point and sufficient top on the inside of vamp, vamp are woven using numerical control weaving is 0.1cm.
Embodiment 20:
Adopt and make customization footwear with the following method:The sensor that user's foot is put on the socks for being equipped with sensor, socks will
The deformation data of foot diverse location is transferred to data processing platform (DPP);Deformation data is changed into picture number using computerized algorithm
According to using CCD optical modeling techniques structure foot threedimensional model, foot threedimensional model is consistent with true foot;Do not make
Shoe tree, numerical control cutting equipment is inputted by foot three-dimensional modeling data, and shoes are made in numerical control cutting equipment automatic cutting shoe body raw material;
Fitted completely with true vola at the interior bottom of shoes;Sole is put into pressure sensor, acceleration transducer, positioner, police simultaneously
Showing device, wireless communication apparatus, power supply and charging device, temperature regulating device, allow shoes to pass through wireless communication device and mobile phone
Special APP connections, user can realize collection exercise data, gait data, plantar pressure data, regulation foot by mobile phone A PP
The functions such as temperature, the individual positioning of realization, panic alarm.
Embodiment 21:
One kind obtains foot image file using the way of contact, and then makes the mode of customization footwear:User's foot plane of stepping on is saturating
On bright glass, clear glass lower end is provided with moveable vola mark scanning equipment (lower scan instrument), clear glass upper end
Around instep and ankle provided with 2-4 moveable scanning devices (top scanner), top scanner and lower scan instrument knot
Close and obtain complete foot image file;Utilize network collection foot data;Three-dimensional modeling, shape are carried out using the foot data being collected into
Into consistent with true foot or close to consistent foot threedimensional model;Shoes, shoes are made according to foot threedimensional model
Interior bottom fitted completely with true vola (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent), any point on the inside of vamp
With the minimum air line distance < 0.48cm of sufficient top (including toes, instep, carpus etc.).
Embodiment 22:
Adopt and make customization footwear with the following method:User claps to obtain 5 foot pictures using photographing function of mobile phone, and picture includes
The graduated object of reference being placed in beside foot;The foot data collected are transferred to data by user by special mobile phone A PP
Processing platform;View data is converted into primary foot threedimensional model using Solidworks modeling softwares, to primary foot three
Dimension module carries out image compensation, repaired, and obtains final foot threedimensional model, foot threedimensional model and true foot one
Cause;Shoe tree is made according to foot threedimensional model, shoes are made further according to shoe tree, the interior bottom of shoes is fitted completely with true vola,
The minimum air line distance on any point and sufficient top is 0.3cm on the inside of vamp;It is no more than above the interior bottom of shoes provided with multiple height
0.3cm projection, can play massage effect.
Embodiment 23:
Adopt and make customization footwear with the following method:Complete foot 3-D view, 3-D scanning are gathered using spatial digitizer
The network interconnection is used between instrument and terminal data processing platform, spatial digitizer is passed automatically or under manual command by foot data
It is defeated to arrive network data processing platform;View data is converted into foot threedimensional model using Pro/E modeling softwares, it is three-dimensional to foot
Model is corrected, and the image at any place lacks compared with true foot when foot threedimensional model is amplified to true foot size
Maximum gauge is fallen into no more than untill 0.2mm;Shoe tree is made according to foot threedimensional model, shoes are made further according to shoe tree, shoes
Interior bottom is fitted completely with true vola, and the minimum air line distance on any point and sufficient top is 0.2cm on the inside of vamp;The interior bottom of shoes
Fixed additional materials are set up in top, the interior bottom of shoes is not exclusively fitted with true vola, for correcting problem foot (such as interior eight
Foot, outer eight foot, the rear strephexopodia, flat foot, talipes cavus, rear strephenopodia etc.) or for treating alleviation foot affliction (such as plantar fascia
Pain, neuralgia, diabetes etc. at inflammation, nurse tumour, metatarsal);Acceleration transducer is inserted in the interior bottom of shoes.
Embodiment 24:
Adopt and make customization footwear with the following method:User sets one section of film for including foot all images information, utilizes hand
Machine APP is transferred to data processing platform (DPP);Multiframe is selected to include foot all images information from film using computerized algorithm
Picture, build foot threedimensional model using CCD optical modeling techniques, foot threedimensional model is consistent with true foot;No
Shoe tree is made, and uses machine vision to judge that combining automatic carving device makes sole, the interior bottom and true vola of shoes are complete
Laminating;The minimum air line distance that any point and sufficient top on the inside of vamp, vamp are woven using numerical control weaving is 0.2cm.
Embodiment 25:
Adopt and make customization footwear with the following method:Foot image file is gathered, source data is obtained;Utilize network collection source data;Figure
As identification, the image being collected into is changed, every foot image file in source data is transformed into space coordinate and gray scale
The digitized image of equal discretization;Secondly, picture noise is removed, i.e., counting Gaussian noise, Rayleigh using probability density function makes an uproar
After sound, gamma noise, exponential distribution noise, uniformly distributed noise and impulsive noise, pre-filtering is carried out to noise image and made an uproar
Sound variance, the high frequency characteristics further according to noise using mean filter method or median filtering method realizes LPF;Again, figure is set up
As degeneration system, i.e., with mathematical function relationship g (x, y)=h (x, y) × f (x, y)+n (x, y) expression degradation models, wherein x, y
The two-dimensional coordinate point of digitized image is represented, g (x, y) represents degraded image, and h (x, y) represents degenrate function, and f (x, y) represents number
The static two dimensional image of word, n (x, y) represents extraneous noise effect, then confirms degenrate function h's (x, y) with digital modeling method
Computational methods;Further according to the removal situation of h (x, y) result of calculation, noise variance and noise, image degradation model is carried out inverse
To derivation, finally it is completely eliminated or the space of part abatement image is degenerated (fuzzy) and point degeneration (noise), obtain restored image.
Recycle computer to strengthen image, Gray Classification is carried out to foot image file, using histogram equalization method to number of pixels
Many gray levels are spreaded, the gray scale few to number of pixels is compressed, and form image local high-contrast, and each is different
Gray level is transformed into different colours according to mapping function linearly or nonlinearly, makes the details of digitlization foot image file brighter
It is aobvious;Then, digitized image is subjected to Fourier transform, different colors is then transformed into by the mapping function of frequency, make number
The details of word foot image file is more obvious;It is that parameter sets up foot HSV space mould with tone (H), saturation degree (S), lightness (V)
Type, the rgb space for digitizing foot image file is mapped on HSV space model, makes the details of digitlization foot image file brighter
It is aobvious.Recycle computer to split image, i.e., be mutated according to gray scale, color change and texture mutation are judged in image not
Continuity Characteristics, and then the border of foot and shooting background in image is extracted, then foot image file is separated from shooting background
Out, and according to the gray scale of image, shape, texture, color the subregion of non-overlapping copies is divided the image into, enables all subregion
It is enough in structure threedimensional model.Extract or mark again sufficient edge, toes, heel, ankle, arch of foot in foot image file etc. to have aobvious
Gray scale, the picture point of chromatic characteristic are write, corresponding points are searched in other clouds and matching relationship is set up;Adjust the ratio of each image
Relation is consistent the engineer's scale of all images, and view data two-dimensional coordinate then is converted into three-dimensional coordinate.Then to figure
As carrying out the overall situations such as denoising, patch formation model defect (filling-up hole), segmentation topography, picture structure, stereo-picture surface be smooth
Optimization processing, obtains consistent with true foot or close to consistent foot threedimensional model;According to foot threedimensional model system
Make shoes, the interior bottoms of shoes is fitted (i.e. the interior bottom of shoes and the shape in true vola is corresponding consistent), footwear completely with true vola
The minimum air line distance < 0.48cm on any point and sufficient top on the inside of face (including toes, instep, carpus etc.).
Claims (10)
1. a kind of customization footwear preparation method based on image recognition and dimensional Modeling Technology, including:Foot image file is gathered, source is obtained
Data;Utilize network collection source data;Image recognition, i.e., sequentially restored to image using computer, strengthened, being divided
Cut, change into the secondary data for being available for three-dimensional modeling;Three-dimensional modeling, i.e., marked using secondary data progress three-dimensional feature extraction,
Coordinate Conversion and point cloud, model global optimization, form consistent with true foot or three-dimensional close to consistent foot
Model;Shoes are made according to foot threedimensional model, the interior bottoms of shoes is fitted (the i.e. interior bottom of shoes and truly completely with true vola
The shape correspondence in vola is consistent), the minimum straight line of any point and sufficient top on the inside of vamp (including toes, instep, carpus etc.) away from
From < 0.48cm.
2. a kind of customization footwear preparation method based on image recognition and dimensional Modeling Technology, it is characterised in that comprise the following steps:
(1) foot image file is gathered:No less than 2 foot image files are shot using filming apparatus and obtain source data;
(2) network collection foot data are utilized:Using cable network or wireless network by foot Data Collection to data processing
Platform;
(3) image recognition:Image is sequentially restored using computer, strengthens, split, changes into and is available for three-dimensional to build
The secondary data of mould;
(4) three-dimensional modeling:Three-dimensional feature, which is carried out, using secondary data extracts mark, Coordinate Conversion and point cloud, the model overall situation
Optimization, forms consistent with true foot or close to consistent foot threedimensional model;
(5) shoes are made:Shoes are made according to foot threedimensional model, wherein:True foot is put on after shoes, the interior bottom of shoes with
Fit completely (i.e. the interior bottom of shoes is corresponding with the shape in true vola consistent) in true vola.
3. preparation method as claimed in claim 1 or 2, it is characterised in that:
The collection foot image file, includes any combination of one or more kinds of modes in the following manner:Utilize camera
Shoot the foot static images that each foot forms more than 3;Shooting foot formation using camera includes the overall shadow of foot
The video of picture;Complete foot 3-D view is gathered using spatial digitizer;Gathered completely using electromagnetic impulse telemetry
Foot 3-D view;Complete foot 3-D view is gathered using running safety method;Utilize mechanical wave pulse ranging method
The complete foot 3-D view of collection;Complete foot 3-D view is gathered using mechanical wave phase ranging method;
The utilization network collection source data, is that the foot image file that will be collected is transferred to data processing by special mobile phone A PP
Platform;Or will be used between the collecting device of foot image file and terminal data processing platform after the network interconnection, each gathered data, adopt
Collect equipment it is automatic or under manual command by foot data transfer to the network terminal;
The source data, the foot image file directly obtained using image capture device, and plantar nervous arch data are turned
The vola view data of chemical conversion;
According to foot threedimensional model, in the following ways in any combination of one or more kinds of modes make shoes:Utilize
3D printing technique directly prints shoes;Make completely the same or close to consistent shoe tree with true foot size and shape, then
Shoes are made using footwear adhesion process;Make completely the same or close to consistent shoe tree with true foot size and shape, so
Afterwards shoes are made using footwear hot vulcanization process;Make completely the same or close to consistent footwear with true foot size and shape
Last carving, then makes shoes using injection (penetrating) technique;Make and true foot size and shape is completely the same or approaches unanimously
Shoe tree, then makes shoes using mould pressing process;Make completely the same or close to consistent footwear with true foot size and shape
Last carving, then makes shoes using sewing tech;Make completely the same or close to consistent shoe-pad with true foot size and shape,
Shoe-pad is put into footwear shoes are made;According to foot threedimensional model, without making shoe tree, using visual determination (including naked eyes sentence
Disconnected and computer vision judges) make shoes.
4. preparation method as claimed in claim 2, it is characterised in that:
It is described that image is restored using computer, comprise the following steps:First, image conversion is carried out, by source data
Every foot image file is transformed into the digitized image of space coordinate and the equal discretization of gray scale;Secondly, remove picture noise, i.e., it is sharp
With probability density function statistics Gaussian noise, rayleigh noise, gamma noise, exponential distribution noise, uniformly distributed noise and pulse
After noise, pre-filtering is carried out to noise image and obtains noise variance, further according to noise high frequency characteristics using mean filter method or
Median filtering method realizes LPF;Again, set up image degradation system, i.e., with mathematical function relationship g (x, y)=h (x, y) ×
F (x, y)+n (x, y) expresses degradation model, and wherein x, y represents the two-dimensional coordinate point of digitized image, and g (x, y) represents Degenerate Graphs
Picture, h (x, y) represents degenrate function, and f (x, y) represents digitized static two dimensional image, and n (x, y) represents extraneous noise effect,
Degenrate function h (x, y) computational methods are confirmed with digital modeling method again;Finally, according to h (x, y) result of calculation, noise variance
With the removal situation of noise, image degradation model is inversely derived, the space of image is finally completely eliminated or partly cuts down
Degenerate (fuzzy) and point is degenerated (noise), obtain restored image;
It is described that image is strengthened using computer, include any combination of one or more kinds of modes in the following manner:
Gray Classification is carried out to foot image file, the gray level more than number of pixels spreaded using histogram equalization method, individual to pixel
The few gray scale of number is compressed, and forms image local high-contrast;Gray Classification is carried out to foot image file, then each is different grey
Degree level is transformed into different colours according to mapping function linearly or nonlinearly, makes the details of digitlization foot image file brighter
It is aobvious;Digitized image is subjected to Fourier transform, different colors are then transformed into by the mapping function of frequency, makes digitlization foot
The details of portion's image is more obvious;It is that parameter sets up foot HSV space model with tone (H), saturation degree (S), lightness (V), will
The rgb space of digitlization foot image file is mapped on HSV space model, makes the details of digitlization foot image file more obvious;With
Cyan (C), carmetta (M), yellow (Y), black (K) are that parameter sets up foot CMYK spatial models, will digitize foot image file
Rgb space be mapped on CMYK spatial models, make the details of digitlization foot image file more obvious;With illumination (L), red is extremely
The brightness range (a) of green, the brightness range (b) of blueness to yellow sets up foot Lab space model for parameter, by digitlization foot
The rgb space of portion's image is mapped on Lab space model, makes the details of digitlization foot image file more obvious;
It is described that image is split using computer, both including being mutated according to gray scale, color change and texture mutation judgement figure
Discontinuity feature as in, and then the border of foot and shooting background in image is extracted, then foot image file is carried on the back from shooting
Separated in scape, any combination of one or more kinds of processes in following process is included again:It will scheme according to the gray scale of image
Subregion as being divided into non-overlapping copies, makes all subregion can be used in building threedimensional model;According to the shape of image by image
The subregion of non-overlapping copies is divided into, makes all subregion can be used in building threedimensional model;Image is divided according to the texture of image
The subregion of non-overlapping copies is cut into, makes all subregion can be used in building threedimensional model;Divided the image into according to the color of image
Into the subregion of non-overlapping copies, all subregion is set to can be used in building threedimensional model.
5. preparation method as claimed in claim 4, it is characterised in that:
After the completion of described image shift step, image is compressed according to Joint Photographic Experts Group;
The digital modeling method confirms degenrate function h (x, y) computational methods, comprises the following steps:Fourier is carried out to h (x, y)
H (x, y) is obtained after leaf transformation, expression formula H (x, y)=2.72^ (- ρ β (x2+y2) are utilized5/6) to H (x, y) assignment, then to assignment
H (x, y) afterwards obtained after reverse Fourier transform h (x, y) last solution, and wherein ρ represents digitized image local unit
Number of greyscale levels in area, β represents the local picture element density of digitized image;
It is described using computer to image carry out it is enhanced during, to color, whether distortion is not specially required;
The histogram equalization method, is to divide the image into some regions, image is transported by region with histogram equalization
Calculate and travel through all regions;Or some regions are divided the image into, each histogram equalization computing at least covers 2 regions;
The discontinuity feature, be by using gray scale Sobel operators or Roberts operators or granny rag Laplacian operater,
Or the single order such as Preveit operators or Kirsch operators or Second Order Differential Operator are detected.
6. preparation method as claimed in claim 1, it is characterised in that:
The collection foot image file, includes any combination of one or more kinds of modes in the following manner:Foot, which is put on, to be equipped with
The wearable device (such as socks) of sensor, foot image file data are calculated by reading the deformation data of foot diverse location;
Foot is put on after the figuratum wearable device (such as socks) of print, and the pattern on wearable device is deformed upon, by shooting not
Wearable device deformation pattern less than 2 obtains foot data;Foot, deformable material are wrapped up with deformable material
In the built-in sensor for being capable of mark position, calculate foot image file data by reading sensing station and displacement;By foot
Put into liquid or gas, the sensor of mark position is capable of built in liquid, gas, by reading sensing station and displacement meter
Calculate foot image file data;
The three-dimensional feature extracts mark, is to find the obvious characteristic point or profile in picture point cloud, then in other clouds
Search corresponding points simultaneously set up matching relationship, and it is extracted or mark mode includes times of one or more kinds of modes in the following manner
Meaning combination:Extract or mark sufficient edge, toes, heel, ankle, arch of foot in foot image file etc. to have notable gray scale, color special
Property picture point, corresponding points are searched in other clouds and matching relationship is set up;Extract foot curved surface in curvature, normal vector,
The invariant features such as the normal vector angle of point-to-point transmission, corresponding points are searched in other clouds and matching relationship is set up;Mark foot figure
1 points of the distance relation two-by-two as in, corresponding points are searched in other clouds and matching relationship is set up;To be spliced
Two panels point cloud or wherein a piece of cloud are sampled, and then make corresponding points on two panels point cloud by iterative calculation using icp algorithms
To or correspondence point-to-plane distance mean square error it is minimum, and then determine matching relationship;With the curvature of curved surface be characterized a little, with angle,
Distance is constraints, and point cloud matching relation is set up using 3 rotary process between a cloud;
The Coordinate Conversion and point cloud, are that the two-dimensional coordinate value of image data is converted into D coordinates value, and by multiple
Image carries out point cloud formation actual match relation using characteristic point or profile as binding site, forms preliminary three-dimensional model, it is sat
The step of mark conversion, sequentially includes:Transformation of scale, that is, adjusting the proportionate relationship of each image makes the engineer's scale of all images
It is consistent;Two-dimensional coordinate is converted into three-dimensional coordinate;
The model global optimization, includes any combination of one or more modes in following optimal way:Denoising, repairing mould
Type defect (filling-up hole), segmentation topography, picture structure, stereo-picture surface are smooth.
7. preparation method as claimed in claim 6, it is characterised in that:
The curvature is Gaussian curvature or average curvature;
The three-dimensional feature point is extracted or mark is, it is necessary to reject bad point matching;
It is described that two panels point cloud to be spliced or wherein a piece of cloud are sampled, including uniform sampling method, stochastical sampling method, method
Vectorial sampling method;
The icp algorithms, after iterative calculation is finished, using the unreliable or incompatible corresponding points pair of the way of restraint removal or point
Face, its way of restraint is a kind of the way of restraint based on curved surface features, that is, constrain foot curved surface in curvature, normal vector, 2 points
Between the invariant features such as normal vector angle.
8. preparation method as claimed in claim 1 or 2, it is characterised in that including following additional technical feature at least within it
One:
Using Solidworks, Pro/E, UG, CAD, Adobe Fuse CC, 3Ds Max, SketchUp, Revit or any its
He realizes three-dimensional modeling by 3 d modeling software;
Inserted in the interior bottom of shoes pressure sensor, and/or acceleration transducer, and/or positioner, and/or alarming device,
And/or wireless communication apparatus, and/or power supply and charging device, and/or temperature regulating device;
The interior bottom of shoes is detachable apparatus, it is preferable that the interior bottom of shoes is shoe-pad, it is preferable that set headspace below shoe-pad,
Pad can be inserted in headspace according to the difference in height of human body double lower limb, human body is got on one's shoes after son and ensure that balance,
It can be used for human body to increase;
Fitted completely with true foot top (including toes, instep, heel, carpus etc.) on the inside of vamp (i.e. the inner side of shoes and
The shape correspondence on true foot top is consistent);
Fixed additional materials are set up above the interior bottom of shoes, the interior bottom of shoes is not exclusively fitted with true vola, for rectifying
Direct problem foot (eight foots, outer eight foot, the rear strephexopodia, flat foot, talipes cavus, rear strephenopodia in such as) alleviates foot for treating
Ailing (such as Plantar Fasciitis, nurse tumour, pain, neuralgia, diabetes at metatarsal);
0.5cm projection is no more than above the interior bottom of shoes provided with multiple height, massage effect can be played, it is preferable that be additional
Material is corresponding with the acupuncture point of sole portion or whole;
User is used after shoes, and Consumer's Experience is fed back into data processing platform (DPP), then footwear construction entered according to user's request
Row adjustment;
Multiple closed individual cavities are provided with the interior bottom of shoes, shoes quality can be mitigated, can also be in human body strenuous exercise
Increase cushioning effect;
The interior bottom of shoes is made using wear-resisting, waterproof, with ductility and elastic material, preferably corium, rubber, poly- ammonia
Carbamate (PU), vinyl acetate copolymer (EVA), MD, TPR, TPUE (TPU), polyvinyl chloride
(PVC), TR, imitation leather bottom, BPU, thermoplastic elastic rubber-sole, makrolon (PC) resin, acrylonitrile-butadiene-styrene (ABS) copolymerization
Thing (ABS), butadiene-styrene rubber (SBR), and their arbitrary composition;
The interior bottom of shoes can be sandwich construction, it is preferable that each layer in interior bottom of shoes is made of different materials;
Height heels are no more than 4cm, preferably 1cm;
Foot threedimensional model is corrected, when foot threedimensional model is amplified to true foot size compared with true foot
Untill the image deflects maximum gauge at any place is no more than 0.5mm, preferably 0.1mm;
Defect recognition is carried out to inner sole using machine vision, when defects with diameters is more than 1mm in footwear, using cutting, and/or note
Modeling, and/or bonding, and/or 3D printing method amendment sole defect;
Three-dimensional modeling uses CCD optical modeling techniques;
When obtaining foot image file using filming apparatus, also include foot partially or completely using the foot data of network collection
Length information;
Sensor of the contact surface of shoes and foot provided with multiple detectable deformation or pressure change, can detect that gait change or
Foot hyperblastosis;
The tensile strength at bottom is not less than 8.5kgf/mm in shoes2, it is preferred that the tensile strength at bottom is not less than 10kgf/ in shoes
mm2;
The bending modulus at bottom is not less than 140kgf/mm in shoes2, it is preferred that the bending modulus at bottom is not less than 180kgf/ in shoes
mm2。
9. a kind of customization footwear manufacturing system based on image recognition and dimensional Modeling Technology, it is characterised in that:The system is wrapped successively
Include:
Data acquisition device, it has equipment of collection foot image file, also with collecting device transmission data to Data Collection
Port;
Data Collection collects device, and it, which has from data acquisition device, receives the ports of data, data storage device, also with to
Pattern recognition device transmits the port of data;
Pattern recognition device, it, which has, collects device from Data Collection and receives the ports of data, image restoration device, image enhaucament
Device, image segmenting device;Also there is the port that data are transmitted to three-dimensional modeling apparatus;
Three-dimensional modeling apparatus, it has the port from pattern recognition device reception data, three-dimensional feature extraction element, Coordinate Conversion
Device, point cloud device, model global optimization device, also with the port that data are transmitted to shoemaking device;
Shoemaking device, it has from the port of three-dimensional modeling apparatus reception data, high-precision vamp making apparatus, high-precision sole
Bottom is fitted (the i.e. interior bottom of shoes and true foot completely with true vola in making apparatus, the shoes that the shoemaking device is produced
The shape correspondence at bottom is consistent), the minimum of any point and sufficient top on the inside of vamp (including toes, instep, heel, carpus etc.) is straight
Linear distance < 0.48cm.
10. manufacturing system as claimed in claim 9, it is characterised in that including following additional technical feature at least one:
The equipment for gathering foot image file, can be wearable device, and/or camera, and/or three-dimensional scanner;
Collect the port that device transmits data to Data Collection, can be mobile phone A PP etc radio port or wired
Port;
The Data Collection collects device, including information pre-processing equipment, for screening effective information and excluding invalid information;
The Data Collection collects device, including DAF, for effective information to be sorted out, and defective data is carried out
Compensation;
The Data Collection collects device, including foot image file instant playback device, and the foot imaging collected can be seen immediately
Data, gathered data can be supplemented at any time when finding data flaw;
The three-dimensional modeling apparatus, including at least level one data converting apparatus, data reforming unit can be by pressure, view data
Change into the computer data for being available for editor;
Described image identifying device and three-dimensional modeling apparatus can be integrated;
The three-dimensional modeling apparatus includes data storage device;
The three-dimensional modeling apparatus includes threedimensional model presentation device, it is preferred that threedimensional model presentation device, which has, arbitrarily to be adjusted
The function and corresponding construction at whole visual angle, and including at least one display screen, can from any angle foot threedimensional model;
The shoemaking device includes the equipment for making shoe tree;
The shoemaking device includes shoe body unit equipment, it uses bonding, sew up, weld, riveting or aforesaid way it is any
Combination realizes the splicing of vamp and sole;
The shoemaking device includes making the equipment and sole-heel splicing apparatus of heel;
The high-precision vamp making apparatus and high-precision sole making apparatus can be integrated in same equipment, it is preferred that high
Precision vamp making apparatus and high-precision sole making apparatus use 3D printing equipment;
The shoemaking device includes visual identity equipment, and it is used for the defect for recognizing inner sole;
The shoemaking device includes inner sole corrective, and it uses cutting, and/or injection, and/or bonding, and/or 3D printing
Mode is used to correct inner sole defect, can be used for correction problem foot (as interior eight foots, outer eight foot, the rear strephexopodia, flat foot,
Talipes cavus, rear strephenopodia etc.) or for treat alleviate foot affliction (such as Plantar Fasciitis, nurse tumour, pain at metatarsal, neuralgia,
Diabetes etc.);
The shoemaking device includes inner sole grinding apparatus, and it is used to polishing smooth inner sole smooth.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710175716.9A CN107095393A (en) | 2017-03-22 | 2017-03-22 | A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710175716.9A CN107095393A (en) | 2017-03-22 | 2017-03-22 | A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107095393A true CN107095393A (en) | 2017-08-29 |
Family
ID=59676210
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710175716.9A Pending CN107095393A (en) | 2017-03-22 | 2017-03-22 | A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107095393A (en) |
Cited By (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107657658A (en) * | 2017-09-26 | 2018-02-02 | 安徽美图信息科技有限公司 | A kind of Virtual Museum display systems being combined based on WEB with threedimensional model |
CN107767375A (en) * | 2017-11-02 | 2018-03-06 | 广东电网有限责任公司电力科学研究院 | One kind point cloud method for evaluating quality and device |
CN107845129A (en) * | 2017-11-07 | 2018-03-27 | 深圳狗尾草智能科技有限公司 | Three-dimensional reconstruction method and device, the method and device of augmented reality |
CN108741428A (en) * | 2018-05-29 | 2018-11-06 | 黑金刚(福建)自动化科技股份公司 | A kind of method of the acquisition identifying system and personalized customization shoes of foot threedimensional model and sufficient vein blending image |
CN108921863A (en) * | 2018-06-12 | 2018-11-30 | 江南大学 | A kind of foot data acquisition device and method |
CN109036548A (en) * | 2018-06-19 | 2018-12-18 | 黑龙江拓盟科技有限公司 | Assisting in diagnosis and treatment system based on mixed reality three-dimensional visualization |
CN109043737A (en) * | 2018-08-13 | 2018-12-21 | 顾萧 | A kind of footwear customization system based on 3-D scanning and three-dimensional modeling |
CN109257543A (en) * | 2018-11-29 | 2019-01-22 | 维沃移动通信有限公司 | Screening-mode control method and mobile terminal |
CN109568123A (en) * | 2018-11-02 | 2019-04-05 | 广东数相智能科技有限公司 | A kind of acupuncture point localization method based on YOLO target detection |
TWI700051B (en) * | 2019-11-07 | 2020-08-01 | 南開科技大學 | Intelligent shoe selection system and method thereof |
CN112231782A (en) * | 2020-06-16 | 2021-01-15 | 北京服装学院 | Shoe product design method based on reverse engineering |
CN114511469A (en) * | 2022-04-06 | 2022-05-17 | 江苏游隼微电子有限公司 | Intelligent image noise reduction prior detection method |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101856162A (en) * | 2009-04-10 | 2010-10-13 | 梦箱舟股份有限公司 | Size-adjustable insole |
CN101964010A (en) * | 2010-09-25 | 2011-02-02 | 四川大学 | Method for designing customized insole |
CN103156331A (en) * | 2011-12-15 | 2013-06-19 | 萨洛蒙股份有限公司 | Footwear with improved tightening of upper |
TWM499093U (en) * | 2014-10-23 | 2015-04-21 | Tai Song Entpr Co Ltd | Vamp adjustment structure of slipper |
CN104657561A (en) * | 2015-03-10 | 2015-05-27 | 南京脚度健康科技有限公司 | Shoe making method applying three-dimensional sole data |
CN104820735A (en) * | 2015-04-17 | 2015-08-05 | 大族激光科技产业集团股份有限公司 | Method and device for making shoes based on three-dimensional reconstruction of feet |
TWM513605U (en) * | 2015-09-03 | 2015-12-11 | Tai Song Entpr Co Ltd | Automatically adjustable vamp |
CN105956349A (en) * | 2016-07-04 | 2016-09-21 | 上海理工大学 | Insole production system based on finite element mechanical analysis |
CN106455757A (en) * | 2014-05-09 | 2017-02-22 | 商用打印机公司 | Methods and apparatuses for designing footwear |
CN106418874A (en) * | 2016-10-13 | 2017-02-22 | 陈茂双 | All-round package, environmental shoes and production process same |
-
2017
- 2017-03-22 CN CN201710175716.9A patent/CN107095393A/en active Pending
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101856162A (en) * | 2009-04-10 | 2010-10-13 | 梦箱舟股份有限公司 | Size-adjustable insole |
CN101964010A (en) * | 2010-09-25 | 2011-02-02 | 四川大学 | Method for designing customized insole |
CN103156331A (en) * | 2011-12-15 | 2013-06-19 | 萨洛蒙股份有限公司 | Footwear with improved tightening of upper |
CN106455757A (en) * | 2014-05-09 | 2017-02-22 | 商用打印机公司 | Methods and apparatuses for designing footwear |
TWM499093U (en) * | 2014-10-23 | 2015-04-21 | Tai Song Entpr Co Ltd | Vamp adjustment structure of slipper |
CN104657561A (en) * | 2015-03-10 | 2015-05-27 | 南京脚度健康科技有限公司 | Shoe making method applying three-dimensional sole data |
CN104820735A (en) * | 2015-04-17 | 2015-08-05 | 大族激光科技产业集团股份有限公司 | Method and device for making shoes based on three-dimensional reconstruction of feet |
TWM513605U (en) * | 2015-09-03 | 2015-12-11 | Tai Song Entpr Co Ltd | Automatically adjustable vamp |
CN105956349A (en) * | 2016-07-04 | 2016-09-21 | 上海理工大学 | Insole production system based on finite element mechanical analysis |
CN106418874A (en) * | 2016-10-13 | 2017-02-22 | 陈茂双 | All-round package, environmental shoes and production process same |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107657658A (en) * | 2017-09-26 | 2018-02-02 | 安徽美图信息科技有限公司 | A kind of Virtual Museum display systems being combined based on WEB with threedimensional model |
CN107767375B (en) * | 2017-11-02 | 2021-06-29 | 广东电网有限责任公司电力科学研究院 | Point cloud quality evaluation method and device |
CN107767375A (en) * | 2017-11-02 | 2018-03-06 | 广东电网有限责任公司电力科学研究院 | One kind point cloud method for evaluating quality and device |
CN107845129A (en) * | 2017-11-07 | 2018-03-27 | 深圳狗尾草智能科技有限公司 | Three-dimensional reconstruction method and device, the method and device of augmented reality |
CN108741428A (en) * | 2018-05-29 | 2018-11-06 | 黑金刚(福建)自动化科技股份公司 | A kind of method of the acquisition identifying system and personalized customization shoes of foot threedimensional model and sufficient vein blending image |
CN108921863A (en) * | 2018-06-12 | 2018-11-30 | 江南大学 | A kind of foot data acquisition device and method |
CN109036548A (en) * | 2018-06-19 | 2018-12-18 | 黑龙江拓盟科技有限公司 | Assisting in diagnosis and treatment system based on mixed reality three-dimensional visualization |
CN109043737A (en) * | 2018-08-13 | 2018-12-21 | 顾萧 | A kind of footwear customization system based on 3-D scanning and three-dimensional modeling |
CN109568123A (en) * | 2018-11-02 | 2019-04-05 | 广东数相智能科技有限公司 | A kind of acupuncture point localization method based on YOLO target detection |
CN109257543A (en) * | 2018-11-29 | 2019-01-22 | 维沃移动通信有限公司 | Screening-mode control method and mobile terminal |
TWI700051B (en) * | 2019-11-07 | 2020-08-01 | 南開科技大學 | Intelligent shoe selection system and method thereof |
CN112231782A (en) * | 2020-06-16 | 2021-01-15 | 北京服装学院 | Shoe product design method based on reverse engineering |
CN114511469A (en) * | 2022-04-06 | 2022-05-17 | 江苏游隼微电子有限公司 | Intelligent image noise reduction prior detection method |
CN114511469B (en) * | 2022-04-06 | 2022-06-21 | 江苏游隼微电子有限公司 | Intelligent image noise reduction prior detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107095393A (en) | A kind of customization footwear preparation method and system based on image recognition and dimensional Modeling Technology | |
CN106942837A (en) | A kind of orthopedics preparation method and system based on image recognition and dimensional Modeling Technology | |
CN107114861A (en) | A kind of customization footwear preparation method and system based on imaging pressure and dimensional Modeling Technology | |
CN106901445A (en) | A kind of orthopedics preparation method and system based on imaging pressure and dimensional Modeling Technology | |
CN107048596A (en) | A kind of preparation method and system for customizing footwear | |
CN106983230B (en) | Method and system for manufacturing orthopedic shoes | |
CN102811640B (en) | An Insole And A Method And A System For Insole Manufacture | |
Wang | An analysis and evaluation of fitness for shoe lasts and human feet | |
CN105243547A (en) | Footwear customization service platform and customization method thereof | |
CN108053283B (en) | Garment customization method based on 3D modeling | |
Luximon et al. | Shoe-last design innovation for better shoe fitting | |
WO2008070537A2 (en) | System and methods of making custom footwear | |
CN110370647B (en) | 3D printing insole and manufacturing method thereof | |
US11176738B2 (en) | Method for calculating the comfort level of footwear | |
CN112449576A (en) | Method and system for obtaining foot analysis data | |
CN111605183A (en) | Manufacturing method and selling method of customized insole | |
CN109043737A (en) | A kind of footwear customization system based on 3-D scanning and three-dimensional modeling | |
IL255924A (en) | Method and machine for preparing insoles | |
KR20220040965A (en) | System for providing foot health customized insole using photography | |
CN114888870B (en) | Insole cutting method for identifying artificial intelligent diabetic foot ulcers based on thermal imaging | |
CN110264304A (en) | The large-scale customization method and system of shoes | |
Mishra et al. | Customization of shoe last based on 3D design process with adjustable 3D ease allowance for better comfort and design | |
WO2021166460A1 (en) | Information provision device, information provision method, and control program | |
WO2020180521A1 (en) | Biometric evaluation of body part images to generate an orthotic | |
US11673321B1 (en) | Cushioned product made using additive manufacture |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information |
Address after: 266000 907 households on the 9th floor of Yulong International Center Building No. 3, 178-2 Haier Road, Laoshan District, Qingdao City, Shandong Province Applicant after: QINGDAO YIXIAOBU TECHNOLOGY CO.,LTD. Address before: No. 177 central Shandong Province Wanda mansion 266000 Qingdao District of Licang City Kyoho Road No. 1 Building 1 unit 902 Applicant before: QINGDAO YIXIAOBU TECHNOLOGY CO.,LTD. |
|
CB02 | Change of applicant information | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170829 |
|
RJ01 | Rejection of invention patent application after publication |