CN110014656A - A kind of 3D printing personalization shoes, print control system and print control program - Google Patents
A kind of 3D printing personalization shoes, print control system and print control program Download PDFInfo
- Publication number
- CN110014656A CN110014656A CN201811524344.7A CN201811524344A CN110014656A CN 110014656 A CN110014656 A CN 110014656A CN 201811524344 A CN201811524344 A CN 201811524344A CN 110014656 A CN110014656 A CN 110014656A
- Authority
- CN
- China
- Prior art keywords
- shoes
- print control
- weight
- printing
- model
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
- B29C64/30—Auxiliary operations or equipment
- B29C64/386—Data acquisition or data processing for additive manufacturing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29C—SHAPING OR JOINING OF PLASTICS; SHAPING OF MATERIAL IN A PLASTIC STATE, NOT OTHERWISE PROVIDED FOR; AFTER-TREATMENT OF THE SHAPED PRODUCTS, e.g. REPAIRING
- B29C64/00—Additive manufacturing, i.e. manufacturing of three-dimensional [3D] objects by additive deposition, additive agglomeration or additive layering, e.g. by 3D printing, stereolithography or selective laser sintering
- B29C64/30—Auxiliary operations or equipment
- B29C64/386—Data acquisition or data processing for additive manufacturing
- B29C64/393—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B33—ADDITIVE MANUFACTURING TECHNOLOGY
- B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- B33Y50/00—Data acquisition or data processing for additive manufacturing
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B33—ADDITIVE MANUFACTURING TECHNOLOGY
- B33Y—ADDITIVE MANUFACTURING, i.e. MANUFACTURING OF THREE-DIMENSIONAL [3-D] OBJECTS BY ADDITIVE DEPOSITION, ADDITIVE AGGLOMERATION OR ADDITIVE LAYERING, e.g. BY 3-D PRINTING, STEREOLITHOGRAPHY OR SELECTIVE LASER SINTERING
- B33Y50/00—Data acquisition or data processing for additive manufacturing
- B33Y50/02—Data acquisition or data processing for additive manufacturing for controlling or regulating additive manufacturing processes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B29—WORKING OF PLASTICS; WORKING OF SUBSTANCES IN A PLASTIC STATE IN GENERAL
- B29L—INDEXING SCHEME ASSOCIATED WITH SUBCLASS B29C, RELATING TO PARTICULAR ARTICLES
- B29L2031/00—Other particular articles
- B29L2031/48—Wearing apparel
- B29L2031/50—Footwear, e.g. shoes or parts thereof
Landscapes
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Materials Engineering (AREA)
- Manufacturing & Machinery (AREA)
- Physics & Mathematics (AREA)
- Mechanical Engineering (AREA)
- Optics & Photonics (AREA)
Abstract
The invention belongs to 3D printing fields, a kind of 3D printing personalization shoes, print control system and print control program are disclosed, pressure distribution when image, the pressure landing of human body step, the temperature of print platform, the movement speed of the extruded velocity of material spray and nozzle are acquired；The relevant information of above-mentioned acquisition is generated into 3D model by the processing of processor；In conjunction with 3D model, corresponding style is selected, is designed integration；Temperature and speed are controlled by nozzle material spray, is printed according to the 3D model of shoes.The present invention is formed using 3D printing, and manufacturing process is simple, and manufacturing speed is high-efficient fastly, can be mass-produced, and can replace appearance, changes pattern, full of individuality.
Description
Technical field
The invention belongs to 3D printing field more particularly to a kind of 3D printing personalization shoes, print control system and printing control
Method processed.
Background technique
Shoes have long development history.In Yangshao culture period about before more than 5000 years, there have been animal skin sewings
Most original shoes.Shoes are a kind of tools that people protect the foot against wound.Earliest people do not allow to overcome special circumstances
Foot is felt bad or injury, has just invented fur shoes.Shoes till now, are formed this present appearance.Various patterns
The shoes of function are seen everywhere., shoes are not simply as the tool walked in people's life, and the shoes having a single function are
Through the requirement for being increasingly difficult to meet people's daily life.Shoes are more complicated in manufacturing process at present, higher cost, and
The type of shoes has many repetitions, with no personalization, and dress is inconvenient, without unique a pair of.
In conclusion problem of the existing technology is:
(1) shoes are more complicated in manufacturing process at present, higher cost, and the type of shoes has many repetitions, does not have
Personalization, dress is inconvenient, without unique a pair of.
(2) pressure sensor measurement step is to the pressure distribution situation on ground, and pressure sensor is using current minimum two
Multiplication carries out error compensation, easily causes the diverging of result.
(3) during velocity sensor acquires the extruded velocity of material spray and the movement speed of nozzle, using existing calculation
Method compensates velocity sensor dynamic, causes velocity sensor to have biggish compensation error, and cannot effectively expansion rate pass
The service band of sensor is not able to satisfy the requirement of the super low-frequency vibration measurement of 3D printing equipment.
(4) picture pick-up device is to the carry out Image Acquisition of step, to image procossing during, using current algorithm pair
Image is handled, and the resolution ratio for obtaining image cannot be effectively enhanced.
Summary of the invention
In view of the problems of the existing technology, the present invention provides a kind of 3D printing personalization shoes, print control system and
Print control program.
The invention is realized in this way a kind of 3D printing personalization shoes print control program, the 3D printing personalization shoes
Print control program includes:
Pressure distribution, the temperature of print platform, the extrusion of material spray speed when the image of acquisition human body step, pressure are landed
The movement speed of degree and nozzle；
The relevant information of above-mentioned acquisition is generated into 3D model by the processing of processor；
In conjunction with 3D model, corresponding style is selected, is designed integration；
Temperature and speed are controlled by nozzle material spray, is printed according to the 3D model of shoes.
Further, in nozzle material spray control speed, the shifting of the extruded velocity and nozzle of velocity sensor acquisition material spray is utilized
Dynamic speed, wherein velocity sensor uses the movement speed of the extruded velocity and nozzle that carry out material spray based on FLANN algorithm to move
State compensation, specifically has:
Input signal u (k) becomes u (k-1) through delay ..., u (k-i) ..., u (k-m) (i=1 ..., m), sensor output
Y (k) becomes through delay
Y (k-1) ..., y (k-j) ..., y (k-n) (j=1 ..., n)；
As the input of network weight adjustment, the input of sensor and the difference e (k) of compensated output u ' (k) are used to
It adjusts weight W (k), after repeatedly training, obtains the inversion model of sensor transfer function difference form, transmission function two
Rank high-pass filter form, is met the requirements using second order compensation inversion model, takes m=2, when n=2, network training formula are as follows:
U ' (k)=W0(k)y(k)+W1(k)y(k-1)+W2(k)y(k-2)+W3(k)u(k)+W4(k)u(k)+W5(k)；
E (k)=u (k)-u ' (k)；
Weighed value adjusting formula is
Wn(k+1)=Wn(k)+ηe(k)y(k-n) (n=0,1,2)；
Wm+2(k+1)=Wm+2(k)+η e (k) u (k-m) (m=0,1,2)；
W5(k+1)=W5(k)+ηe(k)；
In formula: η is Studying factors, and the weight repeatedly obtained after training is the coefficient of inversion model.
Further, using picture pick-up device in the carry out Image Acquisition of step, using being based on Multi-scale model self-similarity
Single image super-resolution algorithms, specifically include:
The initial estimation of image is arranged in step 1Iteration termination error ∈ and maximum number of iterations Kmax；
Step 2 determines down-sampled matrix D and fuzzy matrix H according to the process that degrades of image；
Step 3, establishes image pyramid and as the sample of K-SVD dictionary learning method, passes through dictionary learning
Construct dictionary Ψ；
Step 4 obtains the weight matrix B in NL method by the similar image block of same scale in search image；
Step 5, the current estimation of more new images
Wherein U=(DH)TDH, V=η2(I-B)T(I-B)；
Step 6 updates rarefaction representation coefficient
Wherein, p is the number of image block, and soft (x, τ) is the soft-threshold function of threshold tau,
Soft (x, τ) sgn (x) max (| x |-τ, 0)；
Wherein sgn (x) indicates sign function；
Step 7, the current estimation of more new images
Step 8 repeats step 4~7, carries out next iteration, until iteration result twice in succession meets
Or k >=KmaxIteration ends.
Further, by the processing of processor, 3D model is generated, is specifically included:
1) weight w between neural network model BP hidden layer and output layer is adjustedkj；
Adjust wkjPurpose be desirable to the new output o of output node j* pjO is exported than currentlypjCloser to target value tpj, fixed
Justice:
Wherein α represents the degree of approach, remains unchanged in each cycle of training, and becomes with the adjustment of node in hidden layer H
It is small, do not consider threshold value, then has:
Wherein wkjAnd w* kjRespectively update the weight of front and back, ypkFor hidden layer output, △ wkjFor wkjKnots modification；
According to formulaObtain △ wkjSolution equation:
Wherein,
Equation is solved according to least square and error principleObtain △ wkjApproximate solution:
The hidden layer node k of output node j is connected to each, the weight calculated between k and j changes △ wkj, update
Weight simultaneously calculates error of sum square E, then in one optimal k of k ∈ [1, H] interval selection, so that E is minimum；
2) the weight v between neural network model BP input layer and hidden layer, is adjustedik；
Adjust vikPurpose be once neural network algorithm falls into local minimum point, it is minimum that modification weight can jump out this
Point, judge neural network algorithm fall into local minimum point condition be error E change rate △ E=0, and E > 0；
Do not consider threshold value, the change of the weight of hidden layer node k is solved by following equation:
Wherein δpj=f-1(ypk+Δypk)-f-1(ypk), M is natural number, then hidden layer exports ypkSolution formula are as follows:
Wherein △ ypkFor ypkKnots modification, then have:
According to least square and error principle solution formulaThe matrix equation of building, can
To calculate:
Aggregative formulaWithIt calculates implicit
The consecutive mean knots modification of weight between layer and output layer
Calculate the consecutive mean knots modification of weight between input layer and hidden layer
M takes the natural number between 10~20 in formula, according to formulaWithThe consecutive mean weight for obtaining neural network model BP, according to neural network model BP's
Consecutive mean weight obtains dynamic neural network model DBP；The 3D of personalized shoes is generated according to dynamic neural network model DBP
Model data.
Another object of the present invention is to provide a kind of calculating for realizing the 3D printing personalization shoes print control program
Machine program.
Another object of the present invention is to provide a kind of information for realizing the 3D printing personalization shoes print control program
Data processing terminal.
Another object of the present invention is to provide a kind of computer readable storage mediums, including instruction, when it is in computer
When upper operation, so that computer executes the 3D printing personalization shoes print control program.
Another object of the present invention is to provide a kind of 3D for realizing the 3D printing personalization shoes print control program to beat
Personalized shoes print control system is printed, the 3D printing personalization shoes print control system includes:
Image capture module is connect with central processing module, using picture pick-up device to the carry out Image Acquisition of step；
Pressure acquisition module, connect with central processing module, using pressure sensor measurement step to the pressure point on ground
Cloth situation；
Temperature collecting module is connect with central processing module, using the temperature of temperature sensor acquisition print platform, is prevented
Only the thermal stress distribution of formed shoe is uneven；
Speed acquisition module, connect with central processing module, extruded velocity and spray using velocity sensor acquisition material spray
The movement speed of mouth；
Style selecting module, connect with central processing module, and the style of corresponding shoes is selected by display screen；
Display module is connect with central processing module, shows relevant acquisition information, 3D model, printing using display screen
Procedural information；
3D printing module, connect with central processing module, controls the movement speed of nozzle and the extruded velocity of material spray, into
Row printing.
Another object of the present invention is to provide it is a kind of using the 3D printing personalization shoes print control program printing
3D printing personalization shoes, which is characterized in that the 3D printing personalization shoes are provided with
Shoes cylinder,
The shoes cylinder and vamp are glued, and shoes cylinder rear fluting has connecting sewing；
Vamp shoes cylinder and sole are glued.
Further, magnetic patch is connected on the vamp；Anti-skid chequer is provided on the sole；Connection is installed on connecting sewing
Rope.
Advantages of the present invention and good effect are as follows: the invention is provided with connecting sewing, shoes rear can be opened, can be with
Conveniently wear off；The invention is provided with magnetic patch, and various irony ornaments can be placed on magnetic patch, facilitates the appearance for changing shoes, makes
It has personalization.
The invention is formed using 3D printing, and manufacturing process is simple, and manufacturing speed is high-efficient fastly, can be mass-produced, and
Appearance can be replaced, pattern is changed, it is full of individuality.
Pressure distribution situation in pressure acquisition module of the present invention using pressure sensor measurement step to ground, pressure
Sensor carries out error compensation using current least square method, easily causes the diverging of result, in order to overcome this defect,
Using improved least-squares algorithm.
Speed acquisition module utilizes the extruded velocity of velocity sensor acquisition material spray and the movement speed of nozzle in the present invention
During, it is compensated using the velocity sensor dynamic based on FLANN algorithm, can make velocity sensor that there is smaller benefit
Error is repaid, and effectively extends the service band of velocity sensor, the superlow frequency vibrating for very well satisfying 3D printing equipment is surveyed
The requirement of amount.
In the present invention image capture module using picture pick-up device to the carry out Image Acquisition of step, in order to enhance acquisition figure
The resolution ratio of picture, using the single image super-resolution algorithms based on Multi-scale model self-similarity.
The present invention passes through the processing of processor, generates in 3D model, adjusts neural network model BP hidden layer and output layer
Between weight wkj；Adjust wkjPurpose be desirable to the new output o of output node j* pjO is exported than currentlypjCloser to target value
tpj；Adjust the weight v between neural network model BP input layer and hidden layerik；Deng can accurately obtain the 3D mould of personalized shoes
Type.
Detailed description of the invention
Fig. 1 is 3D printing personalization footwear structure schematic diagram provided in an embodiment of the present invention；
Fig. 2 is 3D printing personalization shoes side schematic view provided in an embodiment of the present invention；
Fig. 3 is 3D printing personalization shoes print control system structural schematic diagram provided in an embodiment of the present invention.
In figure: 1, magnetic patch；2, vamp；3, connecting sewing；4, shoes cylinder；5, connecting rope；6, sole；7, anti-skid chequer；8, image is adopted
Collect module；9, pressure acquisition module；10, temperature collecting module；11, speed acquisition module；12, style selecting module；13, in
Entreat processing module；14, display module；15,3D printing module.
Fig. 4 is 3D printing personalization shoes print control program flow chart provided in an embodiment of the present invention.
Specific embodiment
In order to further understand the content, features and effects of the present invention, the following examples are hereby given, and cooperate attached
Detailed description are as follows for figure.
Structure of the invention is explained in detail with reference to the accompanying drawing.
As Figure 1-Figure 2,3D printing personalization shoes provided in an embodiment of the present invention include: magnetic patch 1, vamp 2, connection
Stitch 3, shoes cylinder 4, connecting rope 5, sole 6, anti-skid chequer 7.
Shoes cylinder 4 and vamp 2 are glued, and 4 rear of shoes cylinder fluting has connecting sewing 3, and 2 shoes cylinder of vamp and sole 6 are glued, on vamp 2
It is connected with magnetic patch 1, anti-skid chequer 7 is provided on sole 6, connecting rope 5 is installed on connecting sewing 3.
The working principle of the invention: the invention prints the sole 6 and vamp 2 and shoes cylinder 4 of shoes, then by 3D printing
6 shoes cylinder 4 of sole is be bonded with vamp 2, anti-skid chequer 7 is printed at 6 bottom of shoes, prevents from skidding, connection is cut into behind shoes cylinder 4
Seam 3, then carries out linking together by connecting rope 5.Magnetic patch 1 is embedded on vamp 2, and irony decoration is placed on vamp 2
Product.
As shown in figure 3,3D printing personalization shoes print control system provided in an embodiment of the present invention:
Image capture module 8 is connect with central processing module 13, using picture pick-up device to the carry out Image Acquisition of step；
Pressure acquisition module 9 is connect with central processing module 13, using pressure sensor measurement step to the pressure on ground
Power distribution situation；
Temperature collecting module 10 is connect with central processing module 13, utilizes the temperature of temperature sensor acquisition print platform
Degree prevents the thermal stress distribution of formed shoe uneven；
Speed acquisition module 11 is connect with central processing module 13, utilizes the extruded velocity of velocity sensor acquisition material spray
With the movement speed of nozzle；
Style selecting module 12 is connect with central processing module 13, and the style of corresponding shoes is selected by display screen；
Display module 14 is connect with central processing module 13, using display screen show relevant acquisition information, 3D model,
Print procedure information；
3D printing module 15 is connect with central processing module 13, controls the movement speed of nozzle and the extrusion speed of material spray
Degree, is printed.
The movement speed of the extruded velocity and nozzle of the utilization of the speed acquisition module 11 velocity sensor acquisition material spray
In the process, in order to which velocity sensor has smaller compensation error, and the service band of velocity sensor is effectively extended, very
The requirement for meeting the super low-frequency vibration measurement of 3D printing equipment well, using the velocity sensor dynamic based on FLANN algorithm
Compensation, detailed process is as follows:
Input signal u (k) becomes u (k-1) through delay ..., u (k-i) ..., u (k-m) (i=1 ..., m), sensor output
Y (k) becomes y (k-1) through delay ...,
Y (k-j) ..., y (k-n) (j=1 ..., n) are using them as the input of network weight adjustment, the input of sensor
It is used to adjust weight W (k) with the difference e (k) of compensated output u ' (k),
After repeatedly training, so that it may obtain the inversion model of sensor transfer function difference form, the magnetic used to application
Electric-type velocity sensor, transmission function are bivalent high-pass filter form, can theoretically be expired using second order compensation inversion model
Foot requires, and takes m=2, when n=2, network training formula are as follows:
U ' (k)=W0(k)y(k)+W1(k)y(k-1)+W2(k)y(k-2)+W3(k)u(k)+W4(k)u(k)+W5(k)；
E (k)=u (k)-u (k)；
Weighed value adjusting formula is
Wn(k+1)=Wn(k)+ηe(k)y(k-n) (n=0,1,2)；
Wm+2(k+1)=Wm+2(k)+η e (k) u (k-m) (m=0,1,2)；
W5(k+1)=W5(k)+ηe(k)；
In formula: η is Studying factors, and the weight repeatedly obtained after training is the coefficient of inversion model.
Described image acquisition module 8, using picture pick-up device to the carry out Image Acquisition of step, in order to enhance acquisition image
Resolution ratio, using the single image super-resolution algorithms based on Multi-scale model self-similarity, specifically includes the following steps:
The initial estimation of image is arranged in step 1Iteration termination error ∈ and maximum number of iterations Kmax；
Step 2 determines down-sampled matrix D and fuzzy matrix H according to the process that degrades of image；
Step 3, establishes image pyramid and as the sample of K-SVD dictionary learning method, passes through dictionary learning
Construct dictionary Ψ；
Step 4 obtains the weight matrix B in NL method by the similar image block of same scale in search image；
Step 5, the current estimation of more new images
Wherein U=(DH)TDH, V=η2(I-B)T(I-B)；
Step 6 updates rarefaction representation coefficient
Wherein, p is the number of image block, and soft (x, τ) is the soft-threshold function of threshold tau,
Soft (x, τ)=sgn (x) max (| x |-τ, 0)；
Wherein sgn (x) indicates sign function；
Step 7, the current estimation of more new images
Step 8 repeats step 4~7, carries out next iteration, until iteration result twice in succession meets
Or k >=KmaxIteration ends.
The pressure acquisition module 9, using pressure sensor measurement step to the pressure distribution situation on ground, pressure is passed
Sensor carries out error compensation using current least square method, easily causes the diverging of result, in order to overcome this defect, adopts
With improved least-squares algorithm, specific calculating process is as follows；
Single input, single Linear Time-Invariant System that exports can use difference equation the following:
A(d-1) y (k)=B (d-1)u(k)；
If it is considered that noise, above formula becomes
A(d-1)(y(k)-ey(k))=B (d-1)(u(k)-eu(k))；
By A (d-1)(y(k)-ey(k))=B (d-1)(u(k)-eu(k)) it writes a Chinese character in simplified form are as follows:
A(d-1)y(k)-B(d-1) u (k)=ε (k)；
U (k) is system input data in formula, and y (k) is output data, d-1For delay operator, ε (k) is regression criterion, and
Have
ε (k)=A (d-1)ey(k)-B(d-1)eu(k)；
Modeling is exactly that the input that obtains according to calibration experiment, output sequence determine model order, and to model parameter ai,
bi, i=0,1,2 ..., n (a0=1) estimation is made, the least square format of model is
In formula.Data vector and parameter vector are set to
Define the target function of Least-squares minimization algorithm:
E=[ε (n+1), ε (n+2) ..., ε (N)] in formulaT。
As shown in figure 4,3D printing personalization shoes print control program provided in an embodiment of the present invention, specifically includes following
Step:
S101: pressure distribution, the temperature of print platform, material spray when the image of acquisition human body step, pressure landing it is crowded
The movement speed of speed and nozzle out；
S102: the relevant information of above-mentioned acquisition is produced into 3D model by the processing of processor；
S103: in conjunction with 3D model, corresponding style is selected, is designed integration；
S104: controlling temperature and speed by nozzle material spray, is printed according to the 3D model of shoes.
By the processing of processor, 3D model is generated, is specifically included:
1) weight w between neural network model BP hidden layer and output layer is adjustedkj；
Adjust wkjPurpose be desirable to the new output o of output node j* pjO is exported than currentlypjCloser to target value tpj, fixed
Justice:
Wherein α represents the degree of approach, remains unchanged in each cycle of training, and becomes with the adjustment of node in hidden layer H
It is small, do not consider threshold value, then has:
Wherein wkjAnd w* kjRespectively update the weight of front and back, ypkFor hidden layer output, △ wkjFor wkjKnots modification；
According to formulaObtain △ wkjSolution equation:
Wherein,
Equation is solved according to least square and error principleObtain △ wkjApproximate solution:
The hidden layer node k of output node j is connected to each, the weight calculated between k and j changes △ wkj, update
Weight simultaneously calculates error of sum square E, then in one optimal k of k ∈ [1, H] interval selection, so that E is minimum；
2) the weight v between neural network model BP input layer and hidden layer, is adjustedik；
Adjust vikPurpose be once neural network algorithm falls into local minimum point, it is minimum that modification weight can jump out this
Point, judge neural network algorithm fall into local minimum point condition be error E change rate △ E=0, and E > 0；
Do not consider threshold value, the change of the weight of hidden layer node k is solved by following equation:
Wherein δpj=f-1(ypk+Δypk)-f-1(ypk), M is natural number, then hidden layer exports ypkSolution formula are as follows:
Wherein △ ypkFor ypkKnots modification, then have:
According to least square and error principle solution formulaThe matrix equation of building, can
To calculate:
Aggregative formulaWithIt calculates implicit
The consecutive mean knots modification of weight between layer and output layer
Calculate the consecutive mean knots modification of weight between input layer and hidden layer
M takes the natural number between 10~20 in formula, according to formulaWithThe consecutive mean weight for obtaining neural network model BP, according to neural network model BP's
Consecutive mean weight obtains dynamic neural network model DBP；The 3D of personalized shoes is generated according to dynamic neural network model DBP
Model data.
In the above-described embodiments, can come wholly or partly by software, hardware, firmware or any combination thereof real
It is existing.When using entirely or partly realizing in the form of a computer program product, the computer program product include one or
Multiple computer instructions.When loading on computers or executing the computer program instructions, entirely or partly generate according to
Process described in the embodiment of the present invention or function.The computer can be general purpose computer, special purpose computer, computer network
Network or other programmable devices.The computer instruction may be stored in a computer readable storage medium, or from one
A computer readable storage medium is transmitted to another computer readable storage medium, for example, the computer instruction can be from
One web-site, computer, server or data center pass through wired (such as coaxial cable, optical fiber, Digital Subscriber Line
(DSL) or wireless (such as infrared, wireless, microwave etc.) mode is into another web-site, computer, server or data
The heart is transmitted).The computer-readable storage medium can be any usable medium that computer can access either
The data storage devices such as server, the data center integrated comprising one or more usable mediums.The usable medium can be
Magnetic medium, (for example, floppy disk, hard disk, tape), optical medium (for example, DVD) or semiconductor medium (such as solid state hard disk
Solid State Disk (SSD)) etc..
The above is only the preferred embodiments of the present invention, and is not intended to limit the present invention in any form,
Any simple modification made to the above embodiment according to the technical essence of the invention, equivalent variations and modification, belong to
In the range of technical solution of the present invention.
Claims (10)
1. a kind of 3D printing personalization shoes print control program, which is characterized in that the 3D printing personalization shoes print control side
Method includes:
Pressure distribution, the temperature of print platform, the extruded velocity of material spray and spray when the image of acquisition human body step, pressure are landed
The movement speed of mouth；
The relevant information of above-mentioned acquisition is generated into 3D model by the processing of processor；
In conjunction with 3D model, corresponding style is selected, is designed integration；
Temperature and speed are controlled by nozzle material spray, is printed according to the 3D model of shoes.
2. 3D printing personalization shoes print control program as described in claim 1, which is characterized in that nozzle material spray controls speed
In, utilize the extruded velocity of velocity sensor acquisition material spray and the movement speed of nozzle, wherein velocity sensor is used and is based on
The movement speed dynamic of extruded velocity and nozzle that FLANN algorithm carries out material spray compensates, and specifically has:
Input signal u (k) becomes u (k-1) through delay ..., u (k-i) ..., u (k-m) (i=1 ..., m), and sensor exports y (k)
Become through delay
Y (k-1) ..., y (k-j) ..., y (k-n) (j=1 ..., n)；
As the input of network weight adjustment, the difference e (k) of the input of sensor and compensated output u ' (k) is used to adjust
Weight W (k) obtains the inversion model of sensor transfer function difference form after repeatedly training, and transmission function is second order high pass
Filter form is met the requirements using second order compensation inversion model, takes m=2, when n=2, network training formula are as follows:
U ' (k)=W0(k)y(k)+W1(k)y(k-1)+W2(k)y(k-2)+W3(k)u(k)+W4(k)u(k)+W5(k)；
E (k)=u (k)-u ' (k)；
Weighed value adjusting formula is
Wn(k+1)=Wn(k)+ηe(k) y (k-n) (n=0,1,2)；
Wm+2(k+1)=Wm+2(k)+ηe(k) u (k-m) (m=0,1,2)；
W5(k+1)=W5(k)+ηe(k)；
In formula: η is Studying factors, and the weight repeatedly obtained after training is the coefficient of inversion model.
3. 3D printing personalization shoes print control program as described in claim 1, which is characterized in that using picture pick-up device to foot
In the carry out Image Acquisition of step, using the single image super-resolution algorithms based on Multi-scale model self-similarity, specifically include:
The initial estimation of image is arranged in step 1Iteration termination error ∈ and maximum number of iterations Kmax；
Step 2 determines down-sampled matrix D and fuzzy matrix H according to the process that degrades of image；
Step 3, establishes image pyramid and as the sample of K-SVD dictionary learning method, constructs word by dictionary learning
Allusion quotation Ψ；
Step 4 obtains the weight matrix B in NL method by the similar image block of same scale in search image；
Step 5, the current estimation of more new images
Wherein U=(DH)TDH, V=η2(I-B)T(I-B)；
Step 6 updates rarefaction representation coefficient
Wherein, p is the number of image block, and soft (χ, τ) is the soft-threshold function of threshold tau,
Soft (x, τ)=sgn (x) max (| x |-τ, 0)；
Wherein sgn (x) indicates sign function；
Step 7, the current estimation of more new images
Step 8 repeats step 4~7, carries out next iteration, until iteration result twice in succession meets
Or k >=KmaxIteration ends.
4. 3D printing personalization shoes print control program as described in claim 1, which is characterized in that by the place of processor
Reason generates 3D model, specifically includes:
1) weight w between neural network model BP hidden layer and output layer is adjustedkj；
Adjust wkjPurpose be desirable to the new output of output node jO is exported than currentlypjCloser to target value tpj, definition:
Wherein α represents the degree of approach, remains unchanged in each cycle of training, and becomes smaller with the adjustment of node in hidden layer H, does not examine
Consider threshold value, then have:
Wherein wkjWithRespectively update the weight of front and back, ypkFor hidden layer output, △ wkjFor wkjKnots modification；
According to formulaObtain △ wkjSolution equation:
Wherein,
Equation is solved according to least square and error principleObtain △ wkjApproximate solution:
The hidden layer node k of output node j is connected to each, the weight calculated between k and j changes △ wkj, update weight
And error of sum square E is calculated, then in one optimal k of k ∈ [1, H] interval selection, so that E is minimum；
2) the weight v between neural network model BP input layer and hidden layer, is adjustedik；
Adjust vikPurpose be once neural network algorithm falls into local minimum point, modification weight can jump out the minimal point, sentence
Disconnected neural network algorithm fall into local minimum point condition be error E change rate △ E=0, and E > 0；
Do not consider threshold value, the change of the weight of hidden layer node k is solved by following equation:
Wherein δpj=f-1(ypk+Δypk)-f-1(ypk), M is natural number, then hidden layer exports ypkSolution formula are as follows:
Wherein △ ypkFor ypkKnots modification, then have:
According to least square and error principle solution formulaThe matrix equation of building can be calculated
Out:
Aggregative formulaWithCalculate hidden layer with
The consecutive mean knots modification of weight between output layer
Calculate the consecutive mean knots modification of weight between input layer and hidden layer
M takes the natural number between 10~20 in formula (12), according to formulaWithThe consecutive mean weight for obtaining neural network model BP, according to neural network model BP's
Consecutive mean weight obtains dynamic neural network model DBP；The 3D mould of personalized shoes is generated according to dynamic neural network model DBP
Type data.
5. a kind of computer program for realizing 3D printing personalization shoes print control program described in Claims 1 to 4 any one.
6. a kind of realize at the information data of 3D printing personalization shoes print control program described in Claims 1 to 4 any one
Manage terminal.
7. a kind of computer readable storage medium, including instruction, when run on a computer, so that computer is executed as weighed
Benefit requires 3D printing personalization shoes print control program described in 1-4 any one.
8. a kind of 3D printing individual character for realizing 3D printing personalization shoes print control program described in Claims 1 to 4 any one
Change shoes print control system, which is characterized in that the 3D printing personalization shoes print control system includes:
Image capture module is connect with central processing module, using picture pick-up device to the carry out Image Acquisition of step；
Pressure acquisition module, connect with central processing module, is distributed feelings using pressure of the pressure sensor measurement step to ground
Condition；
Temperature collecting module is connect with central processing module, using the temperature of temperature sensor acquisition print platform, prevents from forming
The thermal stress distribution of shoes is uneven；
Speed acquisition module, connect with central processing module, utilizes the extruded velocity and nozzle of velocity sensor acquisition material spray
Movement speed；
Style selecting module, connect with central processing module, and the style of corresponding shoes is selected by display screen；
Display module is connect with central processing module, shows relevant acquisition information, 3D model, print procedure using display screen
Information；
3D printing module, connect with central processing module, controls the movement speed of nozzle and the extruded velocity of material spray, is beaten
Print.
9. a kind of 3D printing printed using 3D printing personalization shoes print control program described in 4 any one of Claims 1 to 4
Personalized shoes, which is characterized in that the 3D printing personalization shoes are provided with
Shoes cylinder,
The shoes cylinder and vamp are glued, and shoes cylinder rear fluting has connecting sewing；
Vamp shoes cylinder and sole are glued.
10. 3D printing personalization shoes as claimed in claim 9, which is characterized in that it is connected with magnetic patch on the vamp,；The shoes
Anti-skid chequer is provided on bottom；Connecting rope is installed on connecting sewing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811524344.7A CN110014656A (en) | 2018-12-13 | 2018-12-13 | A kind of 3D printing personalization shoes, print control system and print control program |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811524344.7A CN110014656A (en) | 2018-12-13 | 2018-12-13 | A kind of 3D printing personalization shoes, print control system and print control program |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110014656A true CN110014656A (en) | 2019-07-16 |
Family
ID=67188603
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811524344.7A Pending CN110014656A (en) | 2018-12-13 | 2018-12-13 | A kind of 3D printing personalization shoes, print control system and print control program |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110014656A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110652069A (en) * | 2019-10-17 | 2020-01-07 | 初石智能科技(上海)有限公司 | Extremely-small curved surface structure and manufacturing method thereof |
| CN113103591A (en) * | 2021-04-02 | 2021-07-13 | 苏州斯克莱特数字科技有限公司 | Customized platform for digital 3D printing shoes |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020909A (en) * | 2012-12-06 | 2013-04-03 | 清华大学 | Single-image super-resolution method based on multi-scale structural self-similarity and compressive sensing |
| US20160192741A1 (en) * | 2015-01-05 | 2016-07-07 | Markforged, Inc. | Footwear fabrication by composite filament 3d printing |
| CN106073033A (en) * | 2016-06-08 | 2016-11-09 | 成都市艾米奇鞋业有限公司 | A kind of 3D prints the producing device of shoe tree |
| CN106738871A (en) * | 2015-11-23 | 2017-05-31 | 研能科技股份有限公司 | The method that three-dimensional slippers object is manufactured with fused glass pellet mechanism |
| US20180263329A1 (en) * | 2012-10-22 | 2018-09-20 | Converse Inc. | Customized Shoe Textures and Shoe Portions |
| CN108741393A (en) * | 2012-12-19 | 2018-11-06 | 新平衡运动公司 | Customize footwear and the method for designing and manufacturing it |
| CN108898243A (en) * | 2018-06-04 | 2018-11-27 | 宁德师范学院 | A kind of Transmission Network of Power System security test system |
-
2018
- 2018-12-13 CN CN201811524344.7A patent/CN110014656A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180263329A1 (en) * | 2012-10-22 | 2018-09-20 | Converse Inc. | Customized Shoe Textures and Shoe Portions |
| CN103020909A (en) * | 2012-12-06 | 2013-04-03 | 清华大学 | Single-image super-resolution method based on multi-scale structural self-similarity and compressive sensing |
| CN108741393A (en) * | 2012-12-19 | 2018-11-06 | 新平衡运动公司 | Customize footwear and the method for designing and manufacturing it |
| US20160192741A1 (en) * | 2015-01-05 | 2016-07-07 | Markforged, Inc. | Footwear fabrication by composite filament 3d printing |
| CN106738871A (en) * | 2015-11-23 | 2017-05-31 | 研能科技股份有限公司 | The method that three-dimensional slippers object is manufactured with fused glass pellet mechanism |
| CN106073033A (en) * | 2016-06-08 | 2016-11-09 | 成都市艾米奇鞋业有限公司 | A kind of 3D prints the producing device of shoe tree |
| CN108898243A (en) * | 2018-06-04 | 2018-11-27 | 宁德师范学院 | A kind of Transmission Network of Power System security test system |
Non-Patent Citations (1)
| Title |
|---|
| 石洁等: "基于FLANN算法的速度传感器动态补偿", 《测试技术学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110652069A (en) * | 2019-10-17 | 2020-01-07 | 初石智能科技(上海)有限公司 | Extremely-small curved surface structure and manufacturing method thereof |
| CN113103591A (en) * | 2021-04-02 | 2021-07-13 | 苏州斯克莱特数字科技有限公司 | Customized platform for digital 3D printing shoes |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107808377B (en) | The positioning device of lesion in a kind of lobe of the lung | |
| CN110569886A (en) | Image classification method for bidirectional channel attention element learning | |
| CN105474219A (en) | Systems and methods for estimating physiological heart measurements from medical images and clinical data | |
| CN105809672B (en) | A kind of image multiple target collaboration dividing method constrained based on super-pixel and structuring | |
| CN106407889A (en) | Video human body interaction motion identification method based on optical flow graph depth learning model | |
| CN107180375B (en) | Clothing size recommendation system based on multilayer neural network | |
| Bône et al. | Learning distributions of shape trajectories from longitudinal datasets: a hierarchical model on a manifold of diffeomorphisms | |
| CN110014656A (en) | A kind of 3D printing personalization shoes, print control system and print control program | |
| CN110348579A (en) | A kind of domain-adaptive migration feature method and system | |
| CN110459317A (en) | Alzheimer disease assistant diagnosis system and method based on the dynamic brain network kernel of graph | |
| Mahapatra et al. | Controllable animation of fluid elements in still images | |
| CN108803874A (en) | A kind of human-computer behavior exchange method based on machine vision | |
| CN107862249A (en) | A kind of bifurcated palm grain identification method and device | |
| CN108564012A (en) | A kind of pedestrian's analytic method based on characteristics of human body's distribution | |
| CN108363973A (en) | A kind of unconfined 3D expressions moving method | |
| CN114581356B (en) | Image enhancement model generalization method based on style migration data augmentation | |
| CN116201316B (en) | Close joint paving method and system for large-size ceramic tiles | |
| Zhang et al. | Attributed graph mining and matching: An attempt to define and extract soft attributed patterns | |
| Zeng et al. | TileGAN: category-oriented attention-based high-quality tiled clothes generation from dressed person | |
| CN109902799A (en) | Mix spike neural network and support vector machine classifier | |
| Dietenbeck et al. | Whole myocardium tracking in 2D-echocardiography in multiple orientations using a motion constrained level-set | |
| CN108447101B (en) | A kind of compression of images cognitive method based on JND model | |
| Shah et al. | Walking pattern analysis using deep learning for energy harvesting smart shoes with IoT | |
| CN101571948B (en) | Method for restoring motion blurred image based on total variational model and neutral network | |
| Gang et al. | Skeleton-based action recognition with low-level features of adaptive graph convolutional networks |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190716 |
|
| RJ01 | Rejection of invention patent application after publication |