WO2019157988A1 - 基于3d头部数据的眼镜自动设计***及设计方法 - Google Patents
基于3d头部数据的眼镜自动设计***及设计方法 Download PDFInfo
- Publication number
- WO2019157988A1 WO2019157988A1 PCT/CN2019/074431 CN2019074431W WO2019157988A1 WO 2019157988 A1 WO2019157988 A1 WO 2019157988A1 CN 2019074431 W CN2019074431 W CN 2019074431W WO 2019157988 A1 WO2019157988 A1 WO 2019157988A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- data
- head
- glasses
- design
- spectacle frame
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G02—OPTICS
- G02C—SPECTACLES; SUNGLASSES OR GOGGLES INSOFAR AS THEY HAVE THE SAME FEATURES AS SPECTACLES; CONTACT LENSES
- G02C5/00—Constructions of non-optical parts
-
- G—PHYSICS
- G02—OPTICS
- G02C—SPECTACLES; SUNGLASSES OR GOGGLES INSOFAR AS THEY HAVE THE SAME FEATURES AS SPECTACLES; CONTACT LENSES
- G02C7/00—Optical parts
- G02C7/02—Lenses; Lens systems ; Methods of designing lenses
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Definitions
- the present invention relates to the field of data acquisition and intelligent manufacturing technologies, and in particular, to an automatic eyeglass design system and method based on 3D head data.
- the head data is the key parameter affecting the design of glasses structure. How to accurately collect the head data of Chinese people provides design reference for Chinese glasses design, and solve the glasses frame according to the head data.
- the size should be based on facial morphological features to achieve high comfort becomes an urgent problem to be solved.
- the present invention has been made in order to provide an automatic eyeglass design system based on 3D head data that overcomes the above problems or at least partially solves the above problems.
- An embodiment of the present invention provides an automatic glasses design system based on 3D head data, which includes:
- An image acquisition module configured to acquire a plurality of images of a head, the image acquisition module comprising a camera matrix composed of a plurality of cameras;
- a 3D model generating module configured to extract feature points in the plurality of images, generate point cloud data of the head feature, and generate a 3D model of the head according to the point cloud data to complete collection of the 3D head data;
- a spectacle frame design module configured to extract a design parameter of the spectacle frame according to the 3D head data to form a spectacle frame model
- a lens design module for determining a lens parameter according to a binocular vision condition and a design parameter of the spectacle frame, thereby obtaining a lens model
- the display adjustment module is configured to display the wearing effect of the glasses 3D model, and adjust the spectacle frame model and/or the lens model according to the wearing effect to complete the design of the glasses.
- the 3D model generation module includes:
- a feature point extracting unit configured to process the plurality of images to extract respective feature points in the plurality of images
- a point cloud generating unit configured to perform feature point matching according to the feature of the feature point, establish a matching feature point data set, and calculate a relative position of each camera relative to the head according to optical information of the plurality of cameras Calculating spatial depth information of the feature points in the plurality of images according to the relative position, and the point cloud production unit generates a point cloud of the head feature according to the matched feature point data set and the spatial depth information of the feature points data;
- the 3D model building unit is configured to construct a 3D model of the head according to the point cloud data of the head feature, and complete the collection of the 3D head data.
- the feature point extracting unit uses a scale-invariant feature transform SIFT feature descriptor to describe features of respective feature points in the plurality of images;
- the point cloud generating unit calculates a relative position of each camera relative to the head by using a beam adjustment method
- the spatial depth information includes: spatial location information and color information.
- the 3D model building unit is configured to set a reference size of the 3D model to be constructed, and determine a spatial size of each feature point in the point cloud data according to the reference size and the spatial location information, thereby constructing a 3D header Part model.
- the spectacle frame design module is configured to calculate data of the action area of the spectacle frame according to the 3D head data and extract design parameters of the spectacle frame according to data of the action area of the spectacle frame.
- the data of the action area of the spectacle frame includes: data of the left and right ear screen points and the outer contour of the ear root, the data of the nose and the nose and the bridge of the nose, and the data of the eyelid and the eyelid;
- the design parameters of the spectacle frame include: a temple, a mirror ring, a frame, a bridge of the nose, a nose pad and a pile head.
- the spectacle frame design module is further configured to determine a style of the spectacle frame, and the spectacle frame design module performs size matching calculation on the design parameters of the spectacle frame and the data frame of the style, and forms a spectacle frame according to the calculation result. model.
- the spectacle frame design module scans the existing glasses through a 3D scanner to obtain data of different types of spectacle frames.
- the spectacle frame design module is configured to adjust data of the style spectacle frame according to design parameters of the spectacle frame obtained by the 3D head data, thereby forming a spectacle frame model.
- a 3D printing device for printing the finished glasses according to the glasses 3D model is further included.
- the glasses printed by the 3D printing device are rimless glasses, and the rimless glasses include: a mirror ring, a temple, a bridge of the nose and a nose pad; and the frameless glasses are made of 18K gold.
- the glasses printed by the 3D printing device are half-frame glasses or framed glasses, and the half-frame glasses or the framed glasses all include: a mirror ring, a pile head, a temple, a bridge of the nose and a nose pad; the 3D printing The device prints the mirror ring, temples, pile head and bridge of the nose using a titanium alloy material.
- the image acquisition module further includes:
- the base connected to the support structure is provided with a seat for fixing a photographing position of the human body on the base, and when the human body is located on the seat, the camera head matrix is automatically used to collect the head information of the human body.
- the curved bearing structure is further provided with a display for displaying 3D head data; the display is further configured to set camera parameters of each camera.
- An embodiment of the present invention provides a method for automatically designing glasses based on 3D head data, which includes:
- Step 1 Collecting feature information of the head by using a camera matrix, obtaining multiple images of the head, extracting feature points in the plurality of images, and generating point cloud data of the head feature; constructing a header according to the point cloud data 3D model to achieve 3D head data collection;
- Step 2 Extract the design parameters of the spectacle frame according to the 3D head data to form a 3D model of the spectacle frame;
- Step 3 Lens model: Determine the lens parameters according to the binocular vision condition and the design parameters of the spectacle frame.
- the lens parameters include: type, degree, shape, material and color, and then obtain a 3D model of the lens;
- Step 4. Effect display Display the 3D model of the glasses synthesized according to the spectacle frame model and the lens model, and adjust the spectacle frame model and/or the lens model according to the display effect to complete the glasses design.
- step 1 further includes:
- Point cloud data of the head feature is generated according to the matched feature point data set and the spatial depth information of the feature point.
- step of calculating the relative position of each camera relative to the head according to the optical information of the camera matrix further comprises:
- the relative position of each camera relative to the head is calculated by the beam adjustment method.
- the spatial depth information of the feature points in the plurality of head images includes: spatial position information and color information.
- step of constructing the 3D model of the header according to the point cloud data further includes:
- step 1 the plurality of images are input into a registration algorithm model constructed based on the SIFT algorithm, registration calculation is performed to obtain registration data, and 3D header data is generated based on the registration data.
- step 2 further includes:
- the design parameters of the spectacle frame are extracted according to the data of the action area of the spectacle frame.
- the data of the action area of the spectacle frame includes: data of the left and right ear screen points and the outer contour of the ear root, the data of the nose and the nose and the bridge of the nose, and the data of the eyelid and the eyelid;
- the design parameters of the spectacle frame include: a temple, a mirror ring, a frame, a bridge of the nose, a nose pad and a pile head.
- the step 2 further includes: determining a style of the spectacle frame, performing a size matching calculation on the design parameters of the spectacle frame and the spectacle frame of the style, and forming a 3D model of the spectacle frame according to the calculation result.
- step 2 further comprises: scanning the existing glasses by a 3D scanner to obtain style data of the spectacle frames.
- the size matching calculation includes: adjusting style data of the spectacle frame according to design parameters of the spectacle frame obtained by the 3D head data.
- extracting the design parameters of the spectacle frame according to the data of the action area of the spectacle frame includes:
- step 4 further includes: repeating step 2 and/or step 3 according to the display effect of the glasses 3D model, selecting and determining design parameters and lens parameters of the spectacle frame, and completing the eyeglass design.
- step 4 further comprises: performing the assembled production of the glasses by 3D printing after completing the design of the glasses.
- the glasses are rimless glasses, and the rimless glasses include: a mirror ring, a temple, a bridge of the nose and a nose pad; the frameless glasses are made of 18K gold, and the mirror ring and the mirror are produced by a 3D printer. Legs and bridge of the nose.
- the glasses are half-frame glasses or framed glasses
- the half-frame glasses or the framed glasses include: a mirror ring, a pile head, a temple, a bridge of the nose and a nose pad; the lens ring is produced by a 3D printer, The temple, the pile head and the bridge of the nose; the mirror ring, the temple and the pile head are made of titanium alloy material.
- the invention can achieve the following beneficial effects: the invention automatically designs a system based on 3D head data, adopts multiple cameras to collect feature information of the head, obtains multiple images of the head, and extracts feature points in the plurality of images. Generating point cloud data of the head feature; constructing a 3D model of the head according to the point cloud data to realize acquisition of the 3D head data; extracting design parameters of the spectacle frame according to the 3D head data to form a spectacle frame model; The vision condition and the design parameters of the spectacle frame determine the lens parameters, display the eyeglass model synthesized according to the spectacle frame model and the lens model, and adjust the spectacle frame model and/or the lens model according to the display effect to complete the eyeglass design.
- the assembled production of the glasses is realized by 3D printing, so that the glasses suitable for the head and facial features of the user can be produced quickly, at low cost, and with high matching, thereby improving the user experience.
- the personalization, customization, automatic design and manufacture of the glasses are realized, and the glasses designed and manufactured completely conform to the wearer, thereby improving the comfort.
- the use of multiple camera control technology to collect the head feature information can significantly improve the collection efficiency, and multiple cameras take photos from multiple angles into a complete stereoscopic three-dimensional feature, which can completely restore the head (including the face) in space.
- the various features on the head provide unlimited possibilities for head data applications including design glasses.
- the invention automatically designs a glasses based on 3D head data, adopts a camera matrix to collect feature information of the head, obtains multiple images of the head, extracts feature points in the plurality of images, and generates point cloud data of the head features. Constructing a 3D model of the header according to the point cloud data to implement acquisition of 3D header data; extracting design parameters of the spectacle frame according to the 3D header data to form a 3D model of the spectacle frame; and according to binocular vision conditions and glasses
- the design parameters of the rack determine the lens parameters, display the 3D model of the glasses synthesized according to the spectacle frame model and the lens model, and adjust the spectacle frame model and/or the lens model according to the display effect to complete the eyeglass design.
- the personalization, customization, automatic design and manufacture of the glasses are realized, and the glasses designed and manufactured completely conform to the wearer, thereby improving the comfort.
- the camera matrix control technology is used to collect the head feature information, which can significantly improve the collection efficiency.
- the camera matrix is spliced from multiple angles into a complete stereoscopic three-dimensional feature, which can completely restore the head (including the face) in space.
- the features offer unlimited possibilities for head data applications including design glasses.
- FIG. 1 is a schematic diagram showing internal modules and connections of an automatic eyeglass design system based on 3D head data according to the present invention
- FIG. 2 is a block diagram showing a module flow of an automatic glasses design system based on 3D head data according to another embodiment of the present invention
- FIG. 3 is a schematic diagram of a frame design module for establishing a coordinate system according to collected 3D head data according to an embodiment of the invention
- FIG. 4 is a schematic diagram of a spectacle frame design module determining a temple reference surface based on acquired 3D head data according to an embodiment of the invention
- FIG. 5 is a flowchart showing the operation of an image acquisition module and a 3D model generation module of an automatic glasses design system based on 3D head data according to an embodiment of the invention
- FIG. 6 is a schematic structural view showing a configuration of a camera matrix by arranging a plurality of cameras according to an embodiment of the present invention
- FIG. 7 is a schematic structural diagram of an automatic glasses design system based on 3D head data according to an embodiment of the invention.
- FIG. 8 is a schematic diagram showing a display operation interface of an automatic glasses design system based on 3D head data according to an embodiment of the invention
- FIG. 9 is a diagram showing an internal module of a glasses automatic design system based on 3D head data and a connection diagram thereof according to an embodiment of the present invention.
- an embodiment of the present invention provides an automatic glasses design system based on 3D head data, as shown in FIG. 1 , which includes:
- the image acquisition module 101 is configured to collect multiple images of the head, and the image acquisition module includes a camera matrix composed of a plurality of cameras;
- the 3D model generating module 102 is configured to extract feature points in the plurality of images, generate point cloud data of the head feature, and generate a 3D model of the head according to the point cloud data to complete collection of the 3D head data;
- a spectacle frame design module 103 configured to extract a design parameter of the spectacle frame according to the 3D head data to form a spectacle frame model
- the lens design module 104 is configured to determine a lens parameter according to a binocular vision condition and a design parameter of the spectacle frame, thereby obtaining a lens model;
- the display adjustment module 105 is configured to display the wearing effect of the glasses 3D model, and adjust the frame model and/or the lens model according to the wearing effect to complete the design of the glasses.
- An automatic glasses design system based on 3D head data includes the following operation process:
- the image acquisition module 101 collects feature information of the head to obtain multiple images of the head, and the 3D model generation module 102 extracts feature points in the plurality of images, generates point cloud data of the head feature, and constructs a header according to the point cloud data.
- the spectacle frame design module 103 extracts the design parameters of the spectacle frame according to the 3D head data to form a spectacle frame model, that is, a 3D model of the spectacle frame;
- the lens design module 104 determines the lens parameters according to the binocular vision condition and the design parameters of the spectacle frame.
- the lens parameters include: the type, degree, shape, size, material, color, and the like of the lens, thereby obtaining a lens model, that is, an ophthalmic lens. 3D model;
- the synthetic design module 100 inputs the glasses 3D model to the display adjustment module 105 according to the glasses 3D model synthesized by the spectacle frame model and the lens model, and the user can continuously adjust the spectacle frame model and/or the lens according to the display effect through the display adjustment module 105.
- the model is designed to complete the eyewear, which is based on the best wearing effect. That is, according to the display effect of the glasses 3D model, the design parameters and the lens parameters of the spectacle frame are repeatedly selected and finally determined, and the design of the glasses is completed.
- the invention automatically designs a system based on 3D head data, adopts multiple cameras to collect feature information of the head, obtains multiple images of the head, extracts feature points in the plurality of images, and generates a point cloud of the head features.
- Data constructing a 3D model of the head according to the point cloud data to implement acquisition of 3D head data; extracting design parameters of the spectacle frame according to the 3D head data to form a 3D model of the spectacle frame; according to binocular vision conditions and
- the design parameters of the spectacle frame determine the lens parameters, display the 3D model of the glasses synthesized according to the spectacle frame model and the lens model, and adjust the spectacle frame model and/or the lens model according to the display effect to complete the eyeglass design.
- the personalization, customization, automatic design and manufacture of the glasses are realized.
- the glasses designed and manufactured completely fit the wearer and improve the comfort.
- the use of multiple camera control technologies to collect the head feature information can significantly improve the collection efficiency.
- FIG. 2 is a schematic diagram of a module flow of an automatic glasses design system based on 3D head data according to another embodiment of the present invention, as shown in FIG. 2: S201.
- the user logs in and inputs personal information such as gender, age, height, weight, etc.;
- S202 The head scanning device collects 3D head information of the user, the head scanning device may be the image capturing module 101 above;
- S203 Construct a 3D model of the user's head according to the personal information and the 3D head information; S204. According to the user The 3D model analysis of the head extracts the spectacle frame data suitable for the user; S205.
- the glasses 3D model obtained in the previous step S208 with the user's head model again, adjust the glasses 3D model again and finally determine the glasses 3D data suitable for the user; S210.
- the finalized glasses 3D data in a variety of forms including 3D printing or production data for the production of glasses for assembly production, to obtain customized, personalized, comfortable glasses, can also be virtual wearing experience through video images.
- the glasses automatic design system further includes a 3D printing device, and after the glasses design is completed, the 3D printing device performs printing according to the 3D model of the glasses to realize the assembled production of the glasses, which can be fast, Low-cost, high-matching glasses that are suitable for the user's head and facial features enhance the user experience.
- the glasses may be rimless glasses, framed glasses or half-frame glasses.
- the rimless glasses mainly include: a mirror ring, a temple, a bridge of the nose and a nose pad, wherein the mirror ring, the temple,
- the three components of the bridge of the nose most affect the wearing comfort of the glasses, which is the most important parameter in the design and production of the glasses;
- the mirror ring, temples and bridge of the rimless glasses can be made of 18K gold material, using 18K gold material through the 3D printer production center.
- the framed glasses include: mirror ring, pile head, temple, nose bridge and nose pad.
- mirror ring, temple, pile head and bridge of the nose can be produced by 3D printer.
- the mirror ring, temples, pile head and nose bridge can be made of titanium alloy material with good performance. After 3D printing, these parts can be polished and gold-plated to achieve personalized style and Color design.
- the spectacle frame design module 103 is configured to extract the design parameters of the spectacle frame suitable for the user, that is, the spectacle frame data according to the 3D model analysis of the user's head, and the extracting process includes the following steps: calculating the action area of the spectacle frame according to the 3D head data. Data; and extracting design parameters of the spectacle frame based on data of the area of action of the spectacle frame.
- the data of the action area of the above-mentioned spectacle frame includes data of the left and right ear screen points and the outer contour of the ear root, the data of the nose and the nose and the bridge of the nose, and the data of the eyelid and the eyelid.
- the action area of the glasses frame means that the glasses are placed on the bridge of the nose through the nose pads, and the temples are hung on the ears, and the glasses (glass frames) directly act on the head through the bridge of the nose and the temples, which are embodied in the eyes, the nose, the ears and the surroundings.
- the eyeglass frame does not directly contact the eye, the eyelash and the height of the eye must be considered when customizing the eyeglass.
- the eyelid distance is mainly used to determine the width of the eyeglass lens (frame).
- the height of the pupil that is, the center height of the pupil, refers to the vertical distance from the pupil of the eye to the lowest point of the lower edge of the eyeglass, which is mainly used to determine the height of the spectacle lens (frame).
- the information of the nose mainly affects the customization of the bridge of the nose and the nose pads.
- the glasses are applied to the nose through the bridge of the nose or the nose.
- the width of the bridge of the nose is generally designed to be the width of the nose in a straight line with the eyes.
- the nose pads are located on both sides of the nose and are attached to the nose. Hehe.
- the information of the ear is mainly reflected in the vertical distance from the highest point of the ear to the eye and the height difference from the eye.
- the vertical distance from the highest point of the ear to the eye is used to design the length of the temple.
- the height difference from the highest point of the ear to the eye can be worn.
- due to the weight of the glasses itself it will automatically drop 2-4 mm due to gravity during the wearing process. Therefore, considering the weight factor of the glasses, it is an important factor to realize the customized glasses.
- the data of the action area of the above-mentioned spectacle frame and its influence are considered, and the parameters of the spectacle frame are specifically designed, including: the shape and size of the temple, the mirror ring, the frame, the bridge of the nose, the nose pad and the pile head. , material and other parameters.
- the extraction process may include:
- the spectacle frame design module 103 extracts the design parameters of the spectacle frame from the 3D head data according to the data of the action area of the spectacle frame, and further includes: extracting the outer contour of the ear root: since the temple is applied to the head shape of the human body
- the shape of the ear has an influence on the comfort of the ear and the balance and stability of the whole eye in the head.
- the invention provides a design basis for the design of the ear piece by collecting the contour data of the root of the ear; the face shape extraction of the eye and nose area: the eyeglass frame The nose and the nose are applied to the eye area and the nose area of the face.
- the face shape of the human body is an irregular curved surface, and it is difficult to extract the parameters of the glasses and the nose area, so the marker points are selected in the 3D head model, and Obtain the 3D coordinate value of the marker point to extract the facial parameters, perform principal component analysis on the 3D coordinate value of the marker point, calculate the eye region, the eye frame region and the nose region according to the marker point, and obtain the dip angle of the nose, the nose pad, Nose parameter value; the extraction of the pupil distance, the frame distance design, the distance between the center point of the eye and the frame width and the mirror angle to obtain the pupil distance.
- the spectacle frame design module is further configured to determine the style of the spectacle frame according to the spectacle frame data in the standard spectacle frame database, and perform the size matching calculation on the design parameters of the spectacle frame and the data frame of the style, according to the calculation
- the result is a spectacle frame model.
- the spectacle frame design module scans the existing glasses through the 3D scanner to obtain data of different style frames, and can scan the existing glasses through the 3D scanner to obtain style data of a plurality of different spectacle frames, for example, through multiple visible light cameras.
- the camera matrix is composed of 3D data of the best-selling glasses on the market, and these data are saved to the standard glasses frame database for users to choose, especially for the selection of shapes and styles, and according to the glasses obtained by the user's 3D head data.
- the design parameters of the rack adjust the data corresponding to the style of the glasses frame selected by the user in the standard spectacle frame database, that is, reverse engineering through the 3D scanner, and the best data matching the user is obtained, thereby designing and producing the glasses satisfactory to the user.
- the image acquisition module of the glasses automatic design system based on the 3D head data includes an image acquisition unit 910.
- the image acquisition unit 910 includes a camera matrix composed of a plurality of cameras, and a camera.
- the matrix collects the header information to obtain a plurality of header images.
- the 3D model generation module includes: a feature point extraction unit 920, a point cloud generation unit 930, and a 3D model construction unit 940, where:
- the feature point extraction unit 920 is coupled to the image acquisition unit 910 for processing a plurality of header images to extract respective feature points in the plurality of header images;
- the point cloud generating unit 930 is coupled to the feature point extracting unit 920, and configured to generate feature point cloud data of the head based on the respective feature points in the extracted plurality of head images;
- the 3D model building unit 940 is coupled to the point cloud generating unit 930 for constructing a 3D model of the head according to the feature point cloud data to implement header 3D data collection.
- the point cloud generating unit 930 is further configured to:
- the feature point cloud data of the header is generated according to the matched feature point data set and the spatial depth information of the feature point.
- the features of the respective feature points in the plurality of header images are described using scale invariant feature transform SIFT feature descriptors.
- the point cloud generating unit 930 is further configured to:
- the relative position of each camera relative to the head is calculated by the beam adjustment method.
- the spatial depth information of the feature points in the plurality of head images comprises: spatial position information and color information.
- the foregoing 3D model building unit 940 is further configured to:
- the spatial size of each feature point in the feature point cloud data is determined, thereby constructing a 3D model of the head.
- the 3D model of the head includes at least one of the following 3D data:
- the image acquisition unit 910 of the image acquisition module uses the camera matrix composed of multiple cameras to collect the head information, as shown in FIG. Layout multiple cameras to form a camera matrix:
- the base 31 serves as the main bottom support structure for the entire device
- the display 34 serves as an operation interface and a display interface for the operation of the device system
- the carrying structure 35 is used as a fixed structure of a camera, a central processing unit, and a fill light;
- a band fill light 37 for supplemental use of ambient light is provided.
- the base 31 is connected to the seat 32 through a connecting structure
- the base 31 is connected to the support structure 33 through a mechanism connection structure
- the support structure 33 is connected to the support structure 35 by a mechanical connection structure
- the display 34 is mechanically fixed to the carrying structure 35;
- the camera matrix 36 is fixed on the carrying structure 35 by structural fixing;
- the band fill lamp 37 is fixed to the carrier structure 35 by means of a structural fixing.
- the internal modules of the load bearing structure 35 are composed as follows.
- the internal module of the load bearing structure 35 can be composed of the following parts:
- a power management module that is responsible for providing the various power supplies required for the entire system
- the light management module can adjust the brightness of the light through the central processing module
- the serial port integration module is responsible for two-way communication between the central processing module and the camera matrix
- Central processing module responsible for system information processing, display, lighting, seat control;
- Seat lift management module responsible for seat height adjustment
- the display driver management module is responsible for the display driver of the display.
- the power management module provides power to the camera matrix, the serial port integration module, the light management module, the central processing module, the display drive management module, and the seat lift management module;
- the serial port integration module connects the camera matrix and the central processing module to realize two-way communication between them;
- the camera is connected to the serial port integration module in a single serial manner.
- Serial port integration module is connected to the central processing module via USB interface
- the central processing module realizes the visualization operation of the camera matrix through the customized development software interface
- the camera interface parameters can be set on the operation interface.
- the operation interface can realize the initialization operation of turning on the camera.
- Operation interface can realize the command of camera image acquisition
- the operation interface can realize the setting of camera image storage path
- Operation interface can realize real-time camera browsing and camera switching
- the light management module is connected to the power management module, the central processing module, and the external band fill light;
- the seat lift management module is connected to the power management module, the central processing module and the external seat, and the central processing module realizes the up and down adjustment of the seat height through the visual interface;
- the display driver management module is connected to the power management module, the central processing module, and an external display;
- the central processing module is connected to the power management module, the light management module, the seat lift management module, the serial port integration module, and the display drive management module.
- the device is used as follows
- the startup matrix camera starts to collect information on the human head.
- the information collection time is completed within 0.8 seconds, and the collected signals are finally transmitted to the central processing module in the format of digital image (.jpg).
- the core of the central processing module consists of the following components:
- CPU Central Processing Unit: responsible for the transmission scheduling of the entire digital signal, task allocation, memory management, and some single calculation processing;
- C.2 GPU Graphics Processing Unit: Selects a special model GPU with excellent image processing capabilities and efficient computing capabilities.
- C.3 DRAM Dynamic Random Access Memory
- the signal collected by the matrix camera is transmitted to the central processing module for signal processing.
- D.1 information processing process is as follows
- image filtering can be completed quickly with the support of certain algorithms.
- the format of various information of this device is image format, combined with GPU with excellent image processing capability, the information content of jpg can be evenly distributed to the block of GPU. Since the device uses dual GPUs, each GPU has 56 blocks, so the 18 jpg images captured by the acquisition information are evenly distributed to 112 blocks for calculation, combined with the centralized scheduling and allocation functions of the CPU. It can quickly calculate the feature points of each photo. Compared with the operation of a separate CPU or CPU with other common models of GPUs, the overall operation speed time is 1/10 or less of the latter.
- Image feature points are extracted using the hierarchical structure of the pyramid and the special algorithm of spatial scale invariance. These two special algorithms are combined with the special structure of the GPU selected by the device to maximize the computing performance of the system and achieve fast extraction. Feature points in image information.
- the feature descriptor of this process uses SIFT feature descriptor.
- the SIFT feature descriptor has 128 feature description vectors, which can describe the 128 features of any feature point in direction and scale, and significantly improve the accuracy of feature description.
- the descriptor has spatial independence.
- the special image processing GPU used in this device has excellent calculation and processing capabilities of individual vectors. For SIFT feature vectors with 128 special descriptors, it is most suitable for processing under the conditions of such special GPUs. To take advantage of the special computing power of the GPU, compare the normal CPU or CPU with other common GPUs, and the matching time of feature points will be reduced by 70%.
- the system uses the algorithm of the beam adjustment method to calculate the relative position of the camera relative to the head. According to the spatial coordinates of the relative position, the GPU can quickly calculate the depth information of the head feature points.
- the depth information of the head feature point in space is calculated. Due to the vector computing capability of the GPU, the spatial position and color information of the head feature point cloud can be quickly matched to form a standard model. Cloud information.
- the initial reference size is set for the size of the entire model by the criteria of the feature point cloud size.
- the special calibration has a spatially determined size. Since the head feature point cloud has spatial scale consistency, the size between any feature points of the head can be determined by the size of the special calibration. The spatial position coordinates of the point cloud are calculated.
- the format of 3D data has the following files:
- .mtl - describes the surface material and lighting characteristics of the 3D model
- Head 3D data is displayed on the display by a visual method.
- the camera information is automatically collected by using a camera matrix composed of a plurality of cameras arranged on the curved bearing structure, and after the 3D model of the head is constructed, the head is visually displayed on the display. Part 3D data.
- the camera parameters of each camera are set through the display interface.
- the invention adopts multiple camera control technologies to collect the head information, which can significantly improve the collection efficiency of the head information; and the embodiment of the present invention utilizes the feature information collected in the space of the head to completely restore the head in the space.
- the features on the top provide unlimited possibilities for subsequent application of header data.
- the working process of the 3D model generation module includes the following:
- the generating the feature point cloud data of the header may specifically include the following steps S1061 to S1063.
- Step S1061 Perform matching of the feature points according to the features of the respective feature points in the extracted plurality of head images, and establish a matching feature point data set.
- Step S1062 calculating spatial relative positions of the respective cameras with respect to the head according to the optical information of the plurality of cameras, and calculating spatial depth information of the feature points in the plurality of head images according to the relative positions.
- Step S1063 Generate feature point cloud data of the header according to the matched feature point data set and the spatial depth information of the feature point.
- the features of the respective feature points in the plurality of head images may be described by using SIFT (Scale-Invariant Feature Transform) feature descriptors.
- SIFT Scale-Invariant Feature Transform
- the SIFT feature descriptor has 128 feature description vectors, which can describe the 128 aspects of any feature point in direction and scale, significantly improve the accuracy of feature description, and the feature descriptor has spatial independence.
- the spatial relative position of each camera relative to the head is calculated according to the optical information of the plurality of cameras.
- the embodiment of the present invention provides an optional solution, in which, according to multiple programs, The optical information of the camera is calculated by the beam adjustment method to determine the relative position of each camera relative to the head in space.
- the beam adjustment method is able to extract the coordinates of the 3D point from the multi-view information and The relative position of each camera and the process of optical information.
- the spatial depth information of the feature points in the plurality of head images mentioned in step S1062 may include: spatial position information and color information, that is, may be the X-axis coordinates of the feature points in the spatial position, and the feature points are in the space The Y-axis coordinate of the position, the Z-axis coordinate of the feature point at the spatial position, the value of the R channel of the color information of the feature point, the value of the G channel of the color information of the feature point, the value of the B channel of the color information of the feature point, and the feature The value of the alpha channel of the color information of the point, and so on.
- the generated feature point cloud data includes spatial location information and color information of the feature points, and the format of the feature point cloud data can be as follows:
- Xn represents the X-axis coordinate of the feature point at the spatial position
- Yn represents the Y-axis coordinate of the feature point at the spatial position
- Zn represents the Z-axis coordinate of the feature point at the spatial position
- Rn represents the value of the R-channel of the color information of the feature point.
- Gn represents the value of the G channel of the color information of the feature point
- Bn represents the value of the B channel of the color information of the feature point
- An represents the value of the alpha channel of the color information of the feature point.
- the 3D head data in the embodiment of the present invention refers to that a plurality of cameras take photos taken from multiple angles into a complete stereoscopic three-dimensional feature, and can completely restore various features of the head (including the face) in space.
- the application of subsequent header data provides unlimited possibilities.
- the working process of the 3D model generating module constructing the 3D model of the header according to the feature point cloud data further includes the following: the 3D model building unit of the 3D model generating module sets the reference of the 3D model to be constructed. The size is further determined according to the spatial size information of the reference size and the feature point cloud data, and the spatial size of each feature point in the feature point cloud data is determined, thereby constructing a 3D model of the head.
- 3D data describing the spatial shape feature data of the 3D model, the surface texture feature data describing the 3D model, the surface material describing the 3D model, and the light feature data may be included in the 3D model of the constructed head, which is not used in the embodiment of the present invention. limit.
- FIG. 7 an automatic glasses design system based on 3D head data is shown in FIG. 7, and FIG. 9 is an internal module of the system and a connection diagram thereof, as shown in FIG. 7 and FIG.
- the system includes: a central processing module 701, a camera module 702, an eye degree measurement module 703, a light module 704, a camera rotation mechanism 705, a human body sensing measurement module 706, a camera data transmission module 707, a display operation interface 708, and a base 700.
- the seat 709 can be adjusted.
- the body sensing test module 706 is fixed on the base 700, the display operation interface 708 is connected to the base structure, the central processing module 701 is fixed inside the base, the camera data transmission module 707 is located inside the base structure, and the base 700 is connected to the camera rotation mechanism 705.
- the camera rotating mechanism is connected to the camera module 702, the eye degree measuring module 703 and the light module 704.
- the camera rotating mechanism comprises an adjustable angle camera mount, a rotating device, a rotating device including a servo motor, a speed adjusting box and a transmission device, and a camera module
- the eye degree measuring module and the light module are fixed on the fixed angle fixing frame, and the adjustable angle fixing frame is fixed on the camera rotating mechanism 705, thereby realizing the rotating shooting of the camera with multiple degrees of freedom, and the camera rotating module of the central processing module is connected.
- the servo motor and the adjustable seat 709 are fixed on the base 700, and the adjustable height and the left and right rotation angles of the seat can be adjusted.
- the adjustable seat includes a horizontal rotating servo motor, a vertical lifting servo motor, a horizontal speed adjusting box, and a horizontal rotation. Gear, vertical lift drive gear screw.
- the display operation interface 708 is used to display the wearing effect of the 3D model of the glasses, and input parameters according to the wearing effect to adjust the spectacle frame model and/or the lens model, and finally complete the design of the glasses.
- the seat control module of the central processing module 701 is connected to the adjustable seat.
- the central processing module includes a head image quality processing chip 7011, a 3D model generating module 7012, a glasses frame design module 7019, and a glasses style database 7013.
- the invention automatically designs a system based on 3D head data, adopts multiple cameras to collect feature information of the head, obtains multiple images of the head, extracts feature points in the plurality of images, and generates a point cloud of the head features.
- Data construct a 3D model of the head according to the point cloud data to realize the collection of the 3D head data; extract the design parameters of the spectacle frame according to the 3D head data to form a spectacle frame model; according to the binocular vision condition and the design parameters of the spectacle frame
- the lens parameters are determined, and the glasses model synthesized according to the spectacle frame model and the lens model is displayed, and the spectacle frame model and/or the lens model are adjusted according to the display effect to complete the eyeglass design.
- the assembled production of the glasses is realized by 3D printing, so that the glasses suitable for the head and facial features of the user can be produced quickly, at low cost, and with high matching, thereby improving the user experience.
- the personalization, customization, automatic design and manufacture of the glasses are realized, and the glasses designed and manufactured completely conform to the wearer, thereby improving the comfort.
- the use of multiple camera control technology to collect the head feature information can significantly improve the collection efficiency, and multiple cameras take photos from multiple angles into a complete stereoscopic three-dimensional feature, which can completely restore the head (including the face) in space.
- the various features on the head provide unlimited possibilities for head data applications including design glasses.
- Another embodiment of the present invention provides a method for automatically designing glasses based on 3D head data, which includes the following steps:
- Lens model determining lens parameters according to binocular vision conditions and design parameters of the spectacle frame, the lens parameters include: type, degree, shape, material and color, thereby obtaining a 3D model of the lens;
- S104. Effect display Display the 3D model of the glasses synthesized according to the spectacle frame model and the lens model, and adjust the spectacle frame model and/or the lens model according to the display effect to complete the eyeglass design. That is, step 2 and/or step 3 are repeated according to the display effect of the glasses 3D model, and the design parameters and lens parameters of the spectacle frame are selected and determined, and the eyeglass design is completed.
- the invention automatically designs a glasses based on 3D head data, adopts a camera matrix to collect feature information of the head, obtains multiple images of the head, extracts feature points in the plurality of images, and generates point cloud data of the head features. Constructing a 3D model of the header according to the point cloud data to implement acquisition of 3D header data; extracting design parameters of the spectacle frame according to the 3D header data to form a 3D model of the spectacle frame; and according to binocular vision conditions and glasses
- the design parameters of the rack determine the lens parameters, display the 3D model of the glasses synthesized according to the spectacle frame model and the lens model, and adjust the spectacle frame model and/or the lens model according to the display effect to complete the eyeglass design.
- the glasses are personalized, customized, automated design and manufacture.
- the glasses are designed and manufactured to fit the wearer and improve the comfort.
- the camera matrix control technology is used to collect the head feature information, which can significantly improve the collection efficiency.
- Module flow of the automatic design method of glasses based on 3D head data according to an embodiment of the present invention: S201.
- the user logs in and inputs personal information such as gender, age, height, weight, etc.;
- S202. The head scanning device collects the 3D head of the user Information;
- S203. Constructing a 3D model of the user's head according to the personal information and the 3D header information;
- S204. Extracting the spectacle frame data suitable for the user according to the 3D model analysis of the user's head;
- S205 Measuring the eyes of the user to obtain the binocular vision state According to the vision condition, the lens parameters are obtained, including the type, degree, shape, material and color of the lens, and then the 3D model of the lens is obtained;
- the 3D model of the lens is input into the standard lens data for matching, and the most suitable is obtained.
- Lens S207.
- S208. Get customized glasses 3D data model; S209. It is also possible to compare the 3D model of the glasses obtained in the previous step with the head model of the user again, and to the 3D mode of the glasses. Adjusting and finally determining the 3D data of the glasses suitable for the user; S210.
- the assembled production is performed in various forms including the production data in the form of 3D printing or generating glasses, and customized, personalized Comfortable glasses can also be used for virtual wear through video images.
- the assembled production of the glasses is realized by 3D printing after the design of the glasses is completed, so that the glasses suitable for the head and facial features of the user can be produced quickly, at low cost and with high matching. Improved user experience.
- the glasses may be rimless glasses, framed glasses or half-frame glasses.
- the rimless glasses mainly include: a mirror ring, a temple, a bridge of the nose and a nose pad, wherein the mirror ring, the temple,
- the three components of the bridge of the nose most affect the wearing comfort of the glasses, which is the most important parameter in the design and production of the glasses;
- the mirror ring, temples and bridge of the rimless glasses can be made of 18K gold material, using 18K gold material through the 3D printer production center.
- Mirror ring, temple and nose The 18K gold material can be recycled and reused, so users can trade in the old style when the glasses are outdated or disliked, and the glasses can be changed at will without increasing the cost, further enhancing the experience.
- mirror ring, pile head, temple, nose bridge and nose pad For framed glasses such as half-frame or full-frame, it includes: mirror ring, pile head, temple, nose bridge and nose pad.
- mirror ring, temple, pile head and nose bridge can be produced by 3D printer; Because it is a high-end customized design and manufacture, the mirror ring, temples and pile heads can be made of titanium alloy with good performance. After 3D printing, these parts can be polished and gold-plated to achieve personalized style and color design.
- step S204 extracting the spectacle frame data suitable for the user according to the 3D model analysis of the user's head includes the following steps: calculating data of the action area of the spectacle frame according to the 3D head data; and according to the action area of the spectacle frame The data extracts the design parameters of the spectacle frame.
- the data of the action area of the above-mentioned spectacle frame includes data of the left and right ear screen points and the outer contour of the ear root, data of the dip angle of the nose and nose and the bridge of the nose, and data of the eyelid and the eyelid.
- the action area of the glasses frame means that the glasses are placed on the bridge of the nose through the nose pads, and the temples are hung on the ears, and the glasses (glass frames) directly act on the head through the bridge of the nose and the temples, which are embodied in the eyes, the nose, the ears and the surroundings.
- the eyeglass frame does not directly contact the eye, the eyelash and the height of the eye must be considered when customizing the eyeglass.
- the eyelid distance is mainly used to determine the width of the eyeglass lens (frame).
- the height of the pupil that is, the center height of the pupil, refers to the vertical distance from the pupil of the eye to the lowest point of the lower edge of the eyeglass, which is mainly used to determine the height of the spectacle lens (frame).
- the information of the nose mainly affects the customization of the bridge of the nose and the nose pads.
- the glasses are applied to the nose through the bridge of the nose or the nose.
- the width of the bridge of the nose is generally designed to be the width of the nose in a straight line with the eyes.
- the nose pads are located on both sides of the nose and are attached to the nose. Hehe.
- the information of the ear is mainly reflected in the vertical distance from the highest point of the ear to the eye and the height difference from the eye.
- the vertical distance from the highest point of the ear to the eye is used to design the length of the temple.
- the height difference from the highest point of the ear to the eye can be worn. Provide a basis for the degree of tilt of the temple when the glasses are worn.
- the data of the action area of the above-mentioned spectacle frame and its influence are considered, and the parameters of the spectacle frame are specifically designed, including: the shape and size of the temple, the mirror ring, the frame, the bridge of the nose, the nose pad and the pile head. , material and other parameters.
- extracting the design parameters of the spectacle frame according to the data of the action area of the spectacle frame in step S204 may include:
- step S204 further includes: extracting the outer contour of the ear root: since the temple has an effect on the ear comfort and the balance and stability of the entire eye in the head, in addition to acting on the head shape of the human body, the present invention
- the design of the ear music design is provided by collecting the contour data of the ear root; the facial shape extraction of the eye and nose area: the eyeglass frame and the nose pad act on the eye area and the nose area of the face, but the face shape of the human body is an irregular surface. It is very difficult to extract the parameters of the glasses and the nose area.
- the marker points are selected in the 3D head model, and the 3D coordinate values of the marker points are acquired to extract the facial parameters, and the principal component analysis is performed on the 3D coordinate values of the marker points.
- the method further includes: determining a style of the spectacle frame according to the spectacle frame data in the standard spectacle frame database, performing matching calculation on the design parameters of the spectacle frame and the spectacle frame of the style, and forming according to the calculation result. 3D model of the spectacle frame.
- the existing glasses can be scanned by the 3D scanner to obtain the style data of the spectacle frame.
- the 3D data of the best-selling glasses on the market can be collected by a camera matrix composed of multiple visible light cameras.
- the data is saved to the standard spectacle frame database for user selection, especially for the selection of styling and style, and the eyeglass frame style selected by the user in the standard spectacle frame database is adjusted according to the design parameters of the spectacle frame obtained by the user 3D head data.
- Corresponding data that is, reverse engineering through a 3D scanner, obtains the best data matching the user, and then designs and produces glasses that are satisfactory to the user.
- step S101 of collecting data of the 3D header the method further includes the following steps:
- Point cloud data of the head feature is generated according to the matched feature point data set and the spatial depth information of the feature point.
- the features of the respective feature points in the plurality of head images are described using a scale-invariant feature-converted SIFT feature descriptor.
- the step of calculating the relative position of each camera relative to the head according to the optical information of the camera matrix further comprises:
- the relative position of each camera relative to the head is calculated by the beam adjustment method.
- the spatial depth information of the feature points in the plurality of head images comprises: spatial position information and color information.
- the step of constructing the 3D model of the header according to the point cloud data further comprises:
- the spatial size of each feature point in the feature point cloud data is determined according to the reference size and the spatial position information of the feature point cloud data, thereby constructing a 3D head model.
- Step S101 uses the camera matrix control technology to collect the header information, which can significantly improve the collection efficiency of the header information.
- the embodiment of the present invention utilizes the feature information collected in the space of the head to completely restore the head in space. The various features provide unlimited possibilities for subsequent application of header data.
- the feature point cloud data of the head is generated based on the respective feature points in the extracted plurality of head images in the step S101, and specifically includes the following steps S1061 to S1063. It has been described in detail above and will not be described here.
- modules in the devices of the embodiments can be adaptively changed and placed in one or more devices different from the embodiment.
- the modules or units or components of the embodiments may be combined into one module or unit or component, and further they may be divided into a plurality of sub-modules or sub-units or sub-components.
- any combination of the features disclosed in the specification, including the accompanying claims, the abstract and the drawings, and any methods so disclosed, or All processes or units of the device are combined.
- Each feature disclosed in this specification (including the accompanying claims, the abstract and the drawings) may be replaced by alternative features that provide the same, equivalent or similar purpose.
- the various component embodiments of the present invention may be implemented in hardware, or in a software module running on one or more processors, or in a combination thereof.
- a microprocessor or digital signal processor may be used in practice to implement some or all of the functionality of some or all of the components in accordance with embodiments of the present invention.
- the invention can also be implemented as a device or device program (e.g., a computer program and a computer program product) for performing some or all of the methods described herein.
- a program implementing the invention may be stored on a computer readable medium or may be in the form of one or more signals.
- Such signals may be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Ophthalmology & Optometry (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Optics & Photonics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Eyeglasses (AREA)
Abstract
本发明提供了一种基于3D头部数据的眼镜自动设计***及设计方法,其中,设计***包括用于采集头部多幅图像的图像采集模块;用于提取多幅图像中的特征点的3D模型生成模块;用于根据3D头部数据提取眼镜架的设计参数进而形成眼镜架模型的眼镜架设计模块;用于根据双眼视力状况和眼镜架的设计参数确定镜片参数,进而得到镜片模型的镜片设计模块;用于根据眼镜架模型和镜片模型合成眼镜3D模型的合成设计模块;用于显示眼镜3D模型的佩戴效果,并根据佩戴效果调整眼镜架模型和/或镜片模型以完成眼镜的设计的显示调整模块。本发明实现了眼镜的个性化、定制化、自动化设计和制造,采用多台相机控制技术进行头部特征信息的采集,可以显著提高采集效率和设计准确度。
Description
本发明涉及数据采集和智能制造技术领域,特别涉及一种基于3D头部数据的眼镜自动设计***及方法。
2016年国家***对全国范围内的学校进行学生视力普查数据表明,全国小学生的近视率为27%,初中生为53%,高中生为72%,中专生为50%,大学生为88.48%。然而中国的眼镜市场却存在一种严重的矛盾:一方面中国已经成为了世界上最大的眼镜生产国和眼镜消费国,眼镜的内需市场非常巨大;另一方面,现有的眼镜设计制造却依然简单参照国外眼镜设计方法,设计过程中仍以西方头型特征为主,国内消费者花费高昂费用却依然不能买到一副佩戴舒适的,甚至是适合自己的眼镜。针对中国人的特点设计合适的眼镜有重要意义,头部数据是影响眼镜结构设计的关键参数,如何准确采集国人的头部数据为中国的眼镜设计提供设计参考,并根据头部数据解决眼镜框架的尺寸应基于人脸形态特征以达到高舒适性成为亟待解决的问题。
发明内容
鉴于上述问题,提出了本发明以便提供一种克服上述问题或者至少部分地解决上述问题的一种基于3D头部数据的眼镜自动设计***。
本发明实施例提供了一种基于3D头部数据的眼镜自动设计***,其包括:
图像采集模块,用于采集头部的多幅图像,所述图像采集模块包括由多台相机组成的相机矩阵;
3D模型生成模块,用于提取所述多幅图像中的特征点,生成头部特征的点云数据,并根据所述点云数据生成头部的3D模型,完成3D头部数据的采集;
眼镜架设计模块,用于根据所述3D头部数据提取眼镜架的设计参数进而形成眼镜架模型;
镜片设计模块,用于根据双眼视力状况和所述眼镜架的设计参数确定镜片 参数,进而得到镜片模型;
合成设计模块,用于根据所述眼镜架模型和镜片模型合成的眼镜3D模型;
显示调整模块,用于显示所述眼镜3D模型的佩戴效果,并根据佩戴效果调整眼镜架模型和/或镜片模型以完成眼镜的设计。
进一步的,所述3D模型生成模块包括:
特征点提取单元,用于对所述多幅图像进行处理,提取所述多幅图像中各自的特征点;
点云生成单元,用于根据所述特征点的特征进行特征点的匹配,建立匹配的特征点数据集,根据多台相机的光学信息,计算各台相机相对于头部在空间上的相对位置,根据所述相对位置计算出所述多幅图像中的特征点的空间深度信息,所述点云生产单元根据匹配的特征点数据集和特征点的空间深度信息,生成头部特征的点云数据;
3D模型构建单元,用于根据所述头部特征的点云数据构建头部的3D模型,完成3D头部数据的采集。
进一步的,所述特征点提取单元采用尺度不变特征转换SIFT特征描述子来描述所述多幅图像中各自特征点的特征;
所述点云生成单元采用光束平差法计算各台相机相对于头部在空间上的相对位置;
所述空间深度信息包括:空间位置信息和颜色信息。
进一步的,所述3D模型构建单元用于设定待构建的3D模型的参考尺寸,并根据所述参考尺寸和空间位置信息确定所述点云数据中各个特征点的空间尺寸,从而构建3D头部模型。
进一步的,所述眼镜架设计模块用于根据所述3D头部数据计算眼镜架作用区域的数据并根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数。
进一步的,所述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据;
所述眼镜架的设计参数包括:镜腿、镜圈、镜框、鼻梁、鼻托和桩头。
进一步的,所述眼镜架设计模块还用于确定眼镜架的款式,眼镜架设计模块将所述眼镜架的设计参数与所述款式的眼镜架的数据进行尺寸匹配计算,根据计算结果形成眼镜架模型。
进一步的,所述眼镜架设计模块通过3D扫描仪扫描现有眼镜获得不同款 式眼镜架的数据。
进一步的,所述眼镜架设计模块用于根据由所述3D头部数据获得的眼镜架的设计参数对所述款式眼镜架的数据进行调整,进而形成眼镜架模型。
进一步的,还包括3D打印设备,所述3D打印设备用于根据眼镜3D模型打印完成设计后的眼镜。
进一步的,所述3D打印设备打印的眼镜为无框眼镜,所述无框眼镜包括:镜圈、镜腿、鼻梁和鼻托;所述无框眼镜的材质为18K金。
进一步的,所述3D打印设备打印的眼镜为半框眼镜或有框眼镜,所述半框眼镜或有框眼镜均包括:镜圈、桩头、镜腿、鼻梁和鼻托;所述3D打印设备采用钛合金材料打印所述镜圈、镜腿、桩头和鼻梁。
进一步的,所述图像采集模块还包括:
支撑结构以及设置在所述支撑结构上的弧形承载结构,所述多台相机布置在所述弧形承载结构上形成所述相机矩阵;
与所述支撑结构连接的底座,在所述底座上设有用于固定人体拍照位置的座椅,当人***于所述座椅上时,自动利用所述相机矩阵采集人体的头部信息。
进一步的,所述弧形承载结构上还设有用于显示3D头部数据的显示器;所述显示器还用于设定各台相机的拍照参数。
本发明实施例提供了还一种基于3D头部数据的眼镜自动设计方法,其包括:
步骤1.采用相机矩阵采集头部的特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特征的点云数据;根据所述点云数据构建头部的3D模型,以实现3D头部数据的采集;
步骤2.根据所述3D头部数据提取眼镜架的设计参数,形成眼镜架的3D模型;
步骤3.镜片模型:根据双眼视力状况和眼镜架的设计参数确定镜片参数,镜片参数包括:类型、度数、形状、材质和颜色,进而得到镜片的3D模型;
步骤4.效果显示:显示根据所述眼镜架模型和镜片模型合成的眼镜3D模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。
进一步的,步骤1进一步包括:
根据所述特征点的特征进行特征点的匹配,建立匹配的特征点数据集;
根据相机矩阵的光学信息,计算各台相机相对于头部在空间上的相对位置,并根据所述相对位置计算出所述多幅图像中的特征点的空间深度信息;
根据匹配的特征点数据集和特征点的空间深度信息,生成头部特征的点云数据。
进一步的,所述多幅头部图像中各自的特征点的特征采用尺度不变特征转换SIFT特征描述子来描述。
进一步的,所述根据相机矩阵的光学信息,计算各台相机相对于头部在空间上的相对位置的步骤进一步包括:
根据相机矩阵的光学信息,采用光束平差法计算各台相机相对于头部在空间上的相对位置。
进一步的,所述多幅头部图像中的特征点的空间深度信息包括:空间位置信息和颜色信息。
进一步的,根据点云数据构建头部的3D模型的步骤进一步包括:
设定待构建的3D模型的参考尺寸;
根据所述参考尺寸和所述特征点云数据的空间位置信息,确定所述特征点云数据中各个特征点的空间尺寸,从而构建标定尺寸的3D头部模型。
进一步的,步骤1中:将所述多幅图像输入基于SIFT算法构建的配准算法模型,进行配准计算得到配准数据,基于所述配准数据生成3D头部数据。
进一步的,步骤2进一步包括:
根据所述3D头部数据计算眼镜架作用区域的数据;以及
根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数。
进一步的,所述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据;
所述眼镜架的设计参数包括:镜腿、镜圈、镜框、鼻梁、鼻托和桩头。
进一步的,步骤2进一步包括:确定所述眼镜架的款式,将所述眼镜架的设计参数与所述款式的眼镜架进行尺寸匹配计算,根据计算结果形成眼镜架的3D模型。
进一步的,步骤2进一步包括:通过3D扫描仪扫描现有眼镜,获得所述眼镜架的款式数据。
进一步的,所述尺寸匹配计算包括:根据由所述3D头部数据获得的眼镜架的设计参数,对所述眼镜架的款式数据进行调整。
进一步的,根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数包括:
S21.建立立体三维坐标系:根据3D头部数据,基于左右眼外角点和左右 耳屏点创建平面作为头部数据坐标系的法兰克福面;
S22.镜腿轮廓线的提取:把眼镜腿抽象为一条曲线,作为镜腿参考面,所述镜腿参考面为两眼镜腿的镜脚弯点至镜腿的螺栓中心部分所在的与法兰克福平行的平面;
S23.根据镜腿参考面所在的轮廓设计镜腿弯点长度和镜腿的轮廓。
进一步的,步骤4进一步包括:根据所述眼镜3D模型的显示效果重复步骤2和/或步骤3,选择并确定眼镜架的设计参数和镜片参数,完成眼镜设计。
进一步的,步骤4进一步包括:完成眼镜设计后通过3D打印实现眼镜的装配式生产。
进一步的,所述眼镜为无框眼镜,所述无框眼镜包括:镜圈、镜腿、鼻梁和鼻托;所述无框眼镜的材质为18K金,通过3D打印机生产所述镜圈、镜腿和鼻梁。
进一步的,所述眼镜为半框眼镜或有框眼镜,所述半框眼镜或有框眼镜均包括:镜圈、桩头、镜腿、鼻梁和鼻托;通过3D打印机生产所述镜圈、镜腿、桩头和鼻梁;所述镜圈、镜腿和桩头采用钛合金材料。
本发明能够达到如下有益效果:本发明基于3D头部数据的眼镜自动设计***,采用多台相机采集头部的特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特征的点云数据;根据点云数据构建头部的3D模型,以实现对3D头部数据的采集;根据3D头部数据提取眼镜架的设计参数,形成眼镜架模型;根据双眼视力状况和眼镜架的设计参数确定镜片参数,显示根据所述眼镜架模型和镜片模型合成的眼镜模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。在完成眼镜设计后通过3D打印来实现眼镜的装配式生产,这样可以快速、低成本、高匹配的生产出适合用户头部、面部特征的眼镜,提升了用户体验。实现了眼镜的个性化、定制化、自动化设计和制造,设计制造的眼镜完全贴合佩戴者,提高了舒适度。采用多台相机控制技术进行头部特征信息的采集,可以显著提高采集效率,而且多台相机从多角度拍摄的照片拼接为完整的立体三维特征,可完整地复原头部(包括面部)在空间上的各项特征,为包括设计眼镜在内的头部数据应用提供了无限的可能性。
本发明基于3D头部数据的眼镜自动设计方法,采用相机矩阵采集头部的特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特 征的点云数据;根据所述点云数据构建头部的3D模型,以实现3D头部数据的采集;根据所述3D头部数据提取眼镜架的设计参数,形成眼镜架的3D模型;根据双眼视力状况和眼镜架的设计参数确定镜片参数,显示根据所述眼镜架模型和镜片模型合成的眼镜3D模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。实现了眼镜的个性化、定制化、自动化设计和制造,设计制造的眼镜完全贴合佩戴者,提高了舒适度。采用相机矩阵控制技术进行头部特征信息的采集,可以显著提高采集效率,而且相机矩阵从多角度拍摄的照片拼接为完整的立体三维特征,可完整地复原头部(包括面部)在空间上的各项特征,为包括设计眼镜在内的头部数据应用提供了无限的可能性。
上述说明仅是本发明技术方案的概述,为了能够更清楚了解本发明的技术手段,而可依照说明书的内容予以实施,并且为了让本发明的上述和其它目的、特征和优点能够更明显易懂,以下特举本发明的具体实施方式。
根据下文结合附图对本发明具体实施例的详细描述,本领域技术人员将会更加明了本发明的上述以及其他目的、优点和特征。
通过阅读下文优选实施方式的详细描述,各种其他的优点和益处对于本领域普通技术人员将变得清楚明了。附图仅用于示出优选实施方式的目的,而并不认为是对本发明的限制。而且在整个附图中,用相同的参考符号表示相同的部件。在附图中:
图1示意了根据本发明的基于3D头部数据的眼镜自动设计***的内部模块及连接示意图;
图2示意了根据本发明另一个实施例的基于3D头部数据的眼镜自动设计***的模块流程示意图;
图3示意了根据本发明一实施例的眼镜架设计模块根据采集的3D头部数据建立坐标系的示意图;
图4示意了根据本发明一实施例的眼镜架设计模块根据采集的3D头部数据确定镜腿参考面的示意图;
图5示意了根据本发明一实施例的基于3D头部数据的眼镜自动设计***的图像采集模块及3D模型生成模块的工作流程图;
图6示意了根据本发明一实施例的布局多台相机组成相机矩阵的结构示意 图;
图7示意了根据本发明一实施例的基于3D头部数据的眼镜自动设计***的结构示意图;
图8示意了根据本发明一实施例的基于3D头部数据的眼镜自动设计***的显示操作界面的示意图;
图9示意了根据本发明一实施例的基于3D头部数据的眼镜自动设计***的内部模块及其连接示意图。
下面将参照附图更详细地描述本公开的示例性实施例。虽然附图中显示了本公开的示例性实施例,然而应当理解,可以以各种形式实现本公开而不应被这里阐述的实施例所限制。相反,提供这些实施例是为了能够更透彻地理解本公开,并且能够将本公开的范围完整的传达给本领域的技术人员。
为了解决上述技术问题,本发明实施例提供了一种基于3D头部数据的眼镜自动设计***,如图1所示,其包括:
图像采集模块101,用于采集头部的多幅图像,所述图像采集模块包括由多台相机组成的相机矩阵;
3D模型生成模块102,用于提取所述多幅图像中的特征点,生成头部特征的点云数据,并根据所述点云数据生成头部的3D模型,完成3D头部数据的采集;
眼镜架设计模块103,用于根据所述3D头部数据提取眼镜架的设计参数进而形成眼镜架模型;
镜片设计模块104,用于根据双眼视力状况和所述眼镜架的设计参数确定镜片参数,进而得到镜片模型;
合成设计模块100,用于根据所述眼镜架模型和镜片模型合成的眼镜3D模型;
显示调整模块105,用于显示所述眼镜3D模型的佩戴效果,并根据佩戴效果调整眼镜架模型和/或镜片模型以完成眼镜的设计。
本发明实施例的一种基于3D头部数据的眼镜自动设计***包括如下运行过程:
S101.图像采集模块101采集头部的特征信息,得到头部的多幅图像,3D模型生成模块102提取多幅图像中的特征点,生成头部特征的点云数据并根据 点云数据构建头部的3D模型,以实现3D头部数据的采集;
S102.眼镜架设计模块103根据3D头部数据提取眼镜架的设计参数,形成眼镜架模型,即眼镜架的3D模型;
S103.镜片设计模块104根据双眼视力状况和眼镜架的设计参数确定镜片参数,优选的,镜片参数包括:镜片的类型、度数、形状、尺寸、材质和颜色等,进而得到镜片模型,即眼镜片的3D模型;
S104.合成设计模块100根据眼镜架模型和镜片模型合成的眼镜3D模型,将眼镜3D模型输入到显示调整模块105,用户可通过显示调整模块105根据显示效果不断的调整眼镜架模型和/或镜片模型以完成眼镜设计,以最佳佩戴效果为准。即根据所述眼镜3D模型的显示效果重复选择并最终确定眼镜架的设计参数和镜片参数,完成眼镜的设计。
本发明基于3D头部数据的眼镜自动设计***,采用多台相机采集头部的特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特征的点云数据;根据所述点云数据构建头部的3D模型,以实现3D头部数据的采集;根据所述3D头部数据提取眼镜架的设计参数,形成眼镜架的3D模型;根据双眼视力状况和眼镜架的设计参数确定镜片参数,显示根据所述眼镜架模型和镜片模型合成的眼镜3D模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。实现了眼镜的个性化、定制化、自动化设计和制造,设计制造的眼镜完全贴合佩戴者,提高了舒适度,采用多台相机控制技术进行头部特征信息的采集,可以显著提高采集效率。
图2为根据本发明另一个实施例的基于3D头部数据的眼镜自动设计***的模块流程示意图,如图2所示:S201.用户登陆并输入性别、年龄、身高、体重等个人信息;S202.头部扫描设备采集用户的3D头部信息,该头部扫描设备可以是上文的图像采集模块101;S203.根据个人信息和3D头部信息构建用户头部的3D模型;S204.根据用户头部的3D模型分析提取适合该用户的眼镜架数据;S205.测量用户的双眼,获得双眼视力状况,根据双眼视力状况获得所需的镜片参数,主要包括镜片的类型、度数、形状、材质和颜色等参数,进而得到镜片的3D模型;S206.将镜片的3D模型输入到标准镜片数据中进行匹配,获得最合适的镜片;S207.汇总并分析提取的眼镜架数据、标准眼镜架数据库中的眼镜架数据以及最合适的镜片数据,以满足用户对眼镜的个性化需求;S208.得到定制化的眼镜3D数据模型;S209.将上一步骤S208中获得的眼镜3D模型再次与用户的头部模型相比较,对眼镜3D模型再次调整并最终 确定适合该用户的眼镜3D数据;S210.根据最终确定的眼镜3D数据,采用多种形式包括以3D打印的形式或者生成眼镜的生产数据进行装配式生产,获得定制化、个性化、舒适的眼镜,也可以通过视频影像等形式虚拟佩戴体验。
优选的,在本发明的一个实施例中,该眼镜自动设计***还包括3D打印设备,在完成眼镜设计后通过3D打印设备根据眼镜3D模型进行打印来实现眼镜的装配式生产,这样可以快速、低成本、高匹配的生产出适合用户头部、面部特征的眼镜,提升了用户体验。
优选的,所述眼镜可以是无框眼镜、有框眼镜或半框眼镜,相比有框眼镜,无框眼镜主要包括:镜圈、镜腿、鼻梁和鼻托,其中镜圈、镜腿、鼻梁这三个部件最影响眼镜的佩戴舒适度,为眼镜设计和生产中最重要的参数;无框眼镜的镜圈、镜腿、鼻梁可以采用18K金材料,利用18K金材料通过3D打印机生产所述镜圈、镜腿和鼻梁,18K金材料可以回收再利用,因此用户可以在眼镜款式过时或不喜欢的时候进行以旧换新,无需增加太多成本就可以随意更换眼镜,进一步提升了使用体验。对于半框或全框等有框眼镜而言,有框眼镜均包括:镜圈、桩头、镜腿、鼻梁和鼻托,其中,可以通过3D打印机生产镜圈、镜腿、桩头和鼻梁;优选的,因为是高端定制化设计制造,镜圈、镜腿、桩头和鼻梁可采用性能较好的钛合金材料,3D打印完成后可以对这些部件进行打磨和镀金,实现个性化款式和颜色的设计。
优选的,眼镜架设计模块103用于根据用户头部的3D模型分析提取适合该用户的眼镜架的设计参数即眼镜架数据,其提取过程包括如下步骤:根据3D头部数据计算眼镜架作用区域的数据;以及根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数。上述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据。眼镜架作用区域是指:眼镜通过鼻托放在鼻梁上,镜腿挂在耳朵上,眼镜(眼镜架)通过鼻梁与镜腿直接作用于头部,具体体现在眼睛、鼻子、耳朵及周围。虽然眼镜架不与眼部直接接触,但眼镜定制时必须考虑瞳距和瞳高,以双眼瞳距为例,其主要用作决定眼镜片(镜框)的宽度。瞳高即瞳孔的中心高度,指眼睛瞳孔到眼镜下边缘最低点的垂直距离,其主要用于决定眼镜片(镜框)的高度。鼻子的信息主要影响鼻梁及鼻托的定制,眼镜通过鼻梁或者鼻托作用到鼻子上,鼻梁的宽度一般设计为与双眼在一条直线上鼻子的宽度,鼻托位于鼻子两侧,并与鼻子贴合。耳朵的信息主要表现在耳朵最高点到眼睛的垂直距离以及与眼睛的高度差这两个方面,耳朵最高点到眼睛垂直距离用于设计 镜腿长度,耳朵最高点到眼睛的高度差可以为佩戴眼镜时镜腿的倾斜程度提供依据。另外,由于眼镜自身的重量,在佩戴的过程中会因重力缘故自动下降2-4mm,所以对于眼镜的设计而言,考虑眼镜的重量因素影响是实现定制眼镜的重要因素。
本发明实施例中,考虑了上述眼镜架作用区域的数据及其影响,有针对性的设计眼镜架的参数,包括:镜腿、镜圈、镜框、鼻梁、鼻托和桩头的形状、尺寸、材质等参数。
优选的,为了使眼镜架设计模块103更好的根据眼镜架作用区域的数据从3D头部数据中提取出所述眼镜架的设计参数,该提取过程可以包括:
S21.建立三维坐标系:根据3D头部数据,如图3所示,基于左右眼外角点31和左右耳屏点32创建平面作为头部数据坐标系的法兰克福面30;
S22.镜腿轮廓线的提取:如图4所示,把眼镜腿抽象为一条曲线,作为镜腿参考面40,所述镜腿参考面为两眼镜腿的镜脚弯点至镜腿的螺栓中心部分所在的与法兰克福平行的平面;
S23.根据镜腿参考面所在的轮廓确定镜腿弯点长度和镜腿的轮廓。
优选的,眼镜架设计模块103根据眼镜架作用区域的数据从3D头部数据中提取出所述眼镜架的设计参数还包括:耳根外轮廓的提取:由于镜腿除了作用于人体头型面部外,其耳曲形态对耳部舒适性及整个眼镜在头部的平衡与稳定有影响,本发明通过采集耳朵根部轮廓数据为耳曲设计提供设计依据;眼鼻区域的脸部形态提取:眼镜框与鼻托作用于脸部的眼睛区域与鼻子区域,然而人体的脸部形态为不规则曲面,对眼镜与鼻子区域进行参数提取有很大难度,因此在3D头部模型中选择标记点,并获取标记点的3D坐标值对脸部参数进行提取,对标记点的3D坐标值进行主成分分析,根据标记点,计算眼睛区域、眼框区域和鼻子区域,得到鼻子两侧倾角,鼻托,鼻梁参数值;瞳距的提取,镜框设计时,通过眼睛的中心点距离及镜框宽度和镜面角获取瞳距。
优选的,眼镜架设计模块还用于根据标准眼镜架数据库中的眼镜架数据确定眼镜架的款式,并将上述眼镜架的设计参数与所述款式的眼镜架的数据进行尺寸匹配计算,根据计算结果形成眼镜架模型。
优选的,眼镜架设计模块通过3D扫描仪扫描现有眼镜获得不同款式眼镜架的数据,可以通过3D扫描仪扫描现有眼镜,获得多个不同眼镜架的款式数据,例如可以通过多台可见光相机组成的相机矩阵采集市面上时尚流行畅销的眼镜的3D数据,将这些数据保存至标准眼镜架数据库中供用户选择,尤其是 对造型和款式的选择,并且根据由用户3D头部数据获得的眼镜架的设计参数,调整用户在标准眼镜架数据库中选择的眼镜框架款式对应的数据,即通过3D扫描仪逆向工程,得到匹配该用户的最佳数据,进而设计生产出用户满意的眼镜。
本发明的可选实施例中,基于3D头部数据的眼镜自动设计***的图像采集模块包括图像采集单元910,如图5所示,图像采集单元910包括由多台相机组成的相机矩阵,相机矩阵对头部信息进行采集,得到多幅头部图像;3D模型生成模块包括:特征点提取单元920、点云生成单元930和3D模型构建单元940,其中:
特征点提取单元920,与图像采集单元910相耦合,用于对多幅头部图像进行处理,提取多幅头部图像中各自的特征点;
点云生成单元930,与特征点提取单元920相耦合,用于基于提取的多幅头部图像中各自的特征点,生成头部的特征点云数据;
3D模型构建单元940,与点云生成单元930相耦合,用于根据特征点云数据构建头部的3D模型,以实现头部3D数据的采集。
在本发明的可选实施例中,上述点云生成单元930还用于:
根据提取的多幅头部图像中各自的特征点的特征,进行特征点的匹配,建立匹配的特征点数据集;
根据多台相机的光学信息,计算各台相机相对于头部在空间上的相对位置,并根据相对位置计算出多幅头部图像中的特征点的空间深度信息;
根据匹配的特征点数据集和特征点的空间深度信息,生成头部的特征点云数据。
在本发明的可选实施例中,多幅头部图像中各自的特征点的特征采用尺度不变特征转换SIFT特征描述子来描述。
在本发明的可选实施例中,上述点云生成单元930还用于:
根据多台相机的光学信息,采用光束平差法计算各台相机相对于头部在空间上的相对位置。
在本发明的可选实施例中,多幅头部图像中的特征点的空间深度信息包括:空间位置信息和颜色信息。
在本发明的可选实施例中,上述3D模型构建单元940还用于:
设定待构建的3D模型的参考尺寸;
根据参考尺寸和特征点云数据的空间位置信息,确定特征点云数据中各个 特征点的空间尺寸,从而构建头部的3D模型。
在本发明的可选实施例中,头部的3D模型中包括下列至少之一的3D数据:
描述3D模型的空间形状特征数据;
描述3D模型的表面纹理特征数据;
描述3D模型的表面材质和灯光特征数据。
优选的,为实现图像采集模块快速准确采集头部特征的目的,图像采集模块的图像采集单元910利用多台相机组成的相机矩阵对头部信息进行采集之前,如图6所示,通过以下方式布局多台相机以组成相机矩阵:
搭建支撑结构,在支撑结构上设置弧形承载结构;
将多台相机布置在弧形承载结构上,形成相机矩阵;
搭建与支撑结构连接的底座,在底座上设置用于固定生物拍照位置的座椅;
如图6所示:底座31,作为整个设备的主要的底部支撑结构;
座椅32,用于固定拍照人***置和调节人体高度;
支撑结构33,用于连接设备的底部和其他主体机构;
显示器34,作为设备***工作的操作界面和显示界面;
承载结构35,用作相机、中央处理器、补光灯的固定结构;
相机矩阵36,用于人体头部信息采集;
带状补光灯37,用于环境灯光补充使用。
设备的连接关系说明:
底座31通过连接结构和座椅32相连;
底座31通过机构连接结构和支撑结构33相连;
支撑结构33通过机械连接结构和承载结构35相连;
显示器34通过机械固定在承载结构35上;
相机矩阵36通过结构固定的方式固定在承载结构35上;
带状补光灯37通过结构固定的方式固定在承载结构35上。
承载结构35的内部模块组成如下。
承载结构35的内部模块可以由如下几部分组成:
电源管理模块,负责提供整个***的所需的各种电源;
灯光管理模块,通过中央处理模块可以调整灯光的亮度;
串口集成模块,负责中央处理模块和相机矩阵的双向通讯;
中央处理模块,负责***信息处理、显示、灯光、座椅的控制;
座椅升降管理模块,负责座椅高度调整;
显示驱动管理模块,负责显示器的显示驱动。
承载结构35的内部模块以及外部的连接关系如下:
1)电源管理模块向相机矩阵、串口集成模块、灯光管理模块、中央处理模块、显示驱动管理模块、座椅升降管理模块提供电源;
2)串口集成模块连接相机矩阵和中央处理模块,实现它们之间的双向通讯;
2.1)相机以单独个体的方式以串口的方式和串口集成模块连接
2.2)串口集成模块通过USB接口和中央处理模块连接
2.3)中央处理模块通过定制开发的软件界面实现和相机矩阵的可视化操作
2.4)操作界面上可以实现对相机拍照参数的设置
感光度ISO(范围50~6400)
快门速度(1/4000~1/2)(秒)
变焦倍数(1~3.8x)
光圈(大/小)
2.5)操作界面可以实现对相机开机的初始化操作
2.6)操作界面可以实现相机影像采集的命令
2.7)操作界面可以实现相机影像储存路径的设置
2.8)操作界面可以实现相机实时影像的浏览以及相机的切换
3)灯光管理模块连接电源管理模块、中央处理模块以及外部带状补光灯;
4)座椅升降管理模块连接电源管理模块、中央处理模块以及外部座椅,中央处理模块通过可视化界面实现对座椅高度的上下调节;
5)显示驱动管理模块连接电源管理模块、中央处理模块及外部的显示器;
6)中央处理模块连接电源管理模组、灯光管理模块、座椅升降管理模块、串口集成模块、显示驱动管理模块。
设备使用方法如下
A.启动设备:打开电源开关后,中央处理器,相机矩阵,带状补光灯分别启动。
B.参数设定:通过显示器界面,可以设定相机矩阵拍照的各项参数。
C.信息采集:参数设定完毕后,启动矩阵相机开始对人体头部进行信息采集,信息采集时间在0.8秒内完成,采集的信号最后以数字图像(.jpg)的格 式传至中央处理模块进行处理,中央处理模块核心由以下几个部分组成:
C.1 CPU(Central Processing Unit,中央处理单元):负责整个数字信号的传送调度,任务分配,内存管理,以及部分单一的计算处理;
C.2 GPU(Graphics Processing Unit,图像处理单元):选用特殊型号的GPU,具有优秀的图像处理能力和高效的计算能力;
C.3 DRAM(Dynamic Random Access Memory,即动态随机存取存储器):作为整个数字信号处理的暂时存储中心,需要匹配CPU和GPU的运算能力,得到最佳的处理和计算效能。
D.信息处理:矩阵相机采集完的信号传送到中央处理模块进行信号处理。
D.1信息处理的过程如下
D.1.1采集图像的滤波
利用GPU的特性,结合图像滤波的矩阵运算子的特性,图像滤波可以在一定算法的支持下,快速完成。
D.1.2采集图像的特征点提取
采用CPU和与整体性能相匹配的GPU,因为本设备的各种信息的格式都是图像格式,结合具有优秀图像处理能力的GPU,可以将jpg的各种信息内容均匀的分配到GPU的block中,由于本设备采用双GPU,每颗GPU本身具有56个block,所以采集信息抓取到的18张jpg的图像会均匀的分配到112个block上面进行运算,并结合CPU的集中调度和分配功能,可以快速地计算出每张照片具有的特征点,相对于单独CPU或者CPU搭配其他普通型号的GPU的运算,整体的运算速度时间是后者的1/10或者更短
D.1.3采集图像的匹配和空间深度信息的计算
图像特征点的提取采用金字塔的层级结构,以及空间尺度不变性的特殊算法,这两种特殊的算法都是结合本设备选用的GPU的特殊构造,最大程度的发挥***的计算性能,实现快速提取图像信息中的特征点。
此过程的特征描述子采用SIFT特征描述子,SIFT特征描述子具有128个特征描述向量,可以在方向和尺度上描述任何特征点的128个方面的特征,显著提高对特征描述的精度,同时特征描述子具有空间上的独立性。
本设备采用的特殊图像处理GPU,具有优异的单独向量的计算和处理能力,对于采用128个特殊描述子的SIFT特征向量来讲,在这样特殊GPU的条件下来处理是最适合不过了,可以充分发挥该GPU的特殊计算能力,比较采用普通CPU或者CPU搭配其他普通规格的GPU,特征点的匹配时间会降低 70%。
特征点匹配完毕,***会采用光束平差法的算法计算出相机相对于头部在空间上的相对位置,根据此相对位置的空间坐标,GPU可以快速地计算出头部特征点的深度信息。
D.1.4特征点云数据的生成
根据D.1.3计算出头部特征点在空间的深度信息,由于GPU具有的向量计算能力,可以快速地匹配出头部特征点云的空间位置和颜色信息,形成一个标准的模型建立需要的点云信息。
E.特征尺寸标定:通过特征点云尺寸的标准,为整个模型的尺寸设定最初的参考尺寸。
通过在信息采集上的特殊标定,该特殊标定具有空间确定尺寸,由于头部特征点云具有空间上尺度一致性,通过该特殊标定的确定尺寸,头部的任何特征点之间的尺寸可以从点云的空间位置坐标计算得到。
F.数据的后续处理:基于E中标定的尺寸,通过对点云数据进行进一步的处理,可以得到头部的3D数据。
3D数据的格式有如下几个文件:
.obj——描述3D模型的空间形状特征
.jpg——描述3D模型的表面纹理特征
.mtl——描述3D模型的表面材质和灯光特征
G.头部3D数据通过可视化的方法显示在显示器上。
当人***于座椅上时,利用布置在弧形承载结构上的多台相机组成的相机矩阵自动对头部信息进行采集,在构建得到头部的3D模型后,在显示器上通过可视化方式显示头部3D数据。在利用多台相机组成的相机矩阵对头部信息进行采集之前,通过显示器界面,设定各台相机的拍照参数。
本发明采用多台相机控制技术进行头部信息的采集,可以显著提高头部信息的采集效率;并且,本发明实施例利用采集到头部在空间上的特征信息,完整地复原头部在空间上的各项特征,为后续的头部数据的应用提供了无限的可能性。
在本发明的可选实施例中,3D模型生成模块的工作过程包括如下:
生成头部的特征点云数据,具体可以是包括以下步骤S1061至步骤S1063。
步骤S1061,根据提取的多幅头部图像中各自的特征点的特征,进行特征点的匹配,建立匹配的特征点数据集。
步骤S1062,根据多台相机的光学信息,计算各台相机相对于头部在空间上的相对位置,并根据相对位置计算出多幅头部图像中的特征点的空间深度信息。
步骤S1063,根据匹配的特征点数据集和特征点的空间深度信息,生成头部的特征点云数据。
在上面的步骤S1061中,多幅头部图像中各自的特征点的特征可以采用SIFT(Scale-Invariant Feature Transform,尺度不变特征转换)特征描述子来描述。SIFT特征描述子具有128个特征描述向量,可以在方向和尺度上描述任何特征点的128个方面的特征,显著提高对特征描述的精度,同时特征描述子具有空间上的独立性。
在步骤S1062中,根据多台相机的光学信息,计算各台相机相对于头部在空间上的相对位置,本发明实施例提供了一种可选的方案,在该方案中,可以根据多台相机的光学信息,采用光束平差法计算各台相机相对于头部在空间上的相对位置。
在光束平差法的定义中,假设有一个3D空间中的点,它被位于不同位置的多个相机看到,那么光束平差法就是能够从这些多视角信息中提取出3D点的坐标以及各个相机的相对位置和光学信息的过程。
进一步地,步骤S1062中提及的多幅头部图像中的特征点的空间深度信息可以包括:空间位置信息和颜色信息,即,可以是特征点在空间位置的X轴坐标、特征点在空间位置的Y轴坐标、特征点在空间位置的Z轴坐标、特征点的颜色信息的R通道的值、特征点的颜色信息的G通道的值、特征点的颜色信息的B通道的值、特征点的颜色信息的Alpha通道的值等等。这样,生成的特征点云数据中包含了特征点的空间位置信息和颜色信息,特征点云数据的格式可以如下所示:
X1 Y1 Z1 R1 G1 B1 A1
X2 Y2 Z2 R2 G2 B2 A2
……
Xn Yn Zn Rn Gn Bn An
其中,Xn表示特征点在空间位置的X轴坐标;Yn表示特征点在空间位置的Y轴坐标;Zn表示特征点在空间位置的Z轴坐标;Rn表示特征点的颜色信息的R通道的值;Gn表示特征点的颜色信息的G通道的值;Bn表示特征点的颜色信息的B通道的值;An表示特征点的颜色信息的Alpha通道的值。
在本发明实施例中的3D头部数据,系指多台相机从多角度拍摄的照片拼接为完整的立体三维特征,可完整地复原头部(包括面部)在空间上的各项特征,为后续的头部数据的应用提供了无限的可能性。
在本发明的可选实施例中,3D模型生成模块根据特征点云数据构建头部的3D模型的工作过程还包括如下:3D模型生成模块的3D模型构建单元设定待构建的3D模型的参考尺寸;进而根据参考尺寸和特征点云数据的空间位置信息,确定特征点云数据中各个特征点的空间尺寸,从而构建头部的3D模型。
在构建的头部的3D模型中可以包括描述3D模型的空间形状特征数据、描述3D模型的表面纹理特征数据、描述3D模型的表面材质和灯光特征数据等3D数据,本发明实施例对此不作限制。
进一步的,在本发明的可选实施例中,基于3D头部数据的眼镜自动设计***如图7所示,图9为该***的内部模块及其连接示意图,如图7、图9所示,该***包括:中央处理模块701,相机模块702,眼睛度数测量模块703,灯光模组704,相机旋转机构705,人体感应测量模块706,相机数据传输模块707,显示操作介面708,底座700,可调节座椅709。其中:人体感应测试模块706固定在底座700上,显示操作介面708与底座结构连接,中央处理模块701固定于底座内部,相机数据传输模707块位于底座结构内部,底座700与相机旋转机构705连接,相机旋转机构连接相机模块702、眼睛度数测量模块703和灯光模组704,相机旋转机构包括可调整角度的相机固定架,转动装置,转动装置含伺服电机、配速箱和传动装置,相机模块、眼睛度数测量模块和灯光模组固定在可调整角度的固定架上,可调整角度固定架固定在相机旋转机构705上,实现了相机多自由度的旋转拍摄,中央处理模块的相机转动模块连接伺服电机,可调节座椅709固定于底座700上,可调节座椅上下可调整高度和左右可旋转角度,可调节座椅包括水平转动伺服电机,垂直升降伺服电机,水平配速箱,水平转动齿轮,垂直升降传动齿轮螺杆。如图8所示,显示操作介面708用于显示眼镜3D模型的佩戴效果,并根据佩戴效果输入参数以调整眼镜架模型和/或镜片模型,最终完成眼镜的设计。
如图9所示,中央处理模块701的座椅控制模块连接该可调节座椅,中央处理模块包括头部图像品质处理芯片7011、3D模型生成模块7012、眼镜架设计模块7019、眼镜款式数据库7013、眼镜片款式数据库7014,微处理器控制模块7015,座椅控制模块7017、相机转动控制模块7018和灯光控制模块7016。
本发明基于3D头部数据的眼镜自动设计***,采用多台相机采集头部的 特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特征的点云数据;根据点云数据构建头部的3D模型,以实现对3D头部数据的采集;根据3D头部数据提取眼镜架的设计参数,形成眼镜架模型;根据双眼视力状况和眼镜架的设计参数确定镜片参数,显示根据所述眼镜架模型和镜片模型合成的眼镜模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。在完成眼镜设计后通过3D打印来实现眼镜的装配式生产,这样可以快速、低成本、高匹配的生产出适合用户头部、面部特征的眼镜,提升了用户体验。实现了眼镜的个性化、定制化、自动化设计和制造,设计制造的眼镜完全贴合佩戴者,提高了舒适度。采用多台相机控制技术进行头部特征信息的采集,可以显著提高采集效率,而且多台相机从多角度拍摄的照片拼接为完整的立体三维特征,可完整地复原头部(包括面部)在空间上的各项特征,为包括设计眼镜在内的头部数据应用提供了无限的可能性。
本发明另一实施例提供了还一种基于3D头部数据的眼镜自动设计方法,,其包括如下步骤:
S101.采用相机矩阵采集头部的特征信息,得到头部的多幅图像,提取多幅图像中的特征点,生成头部特征的点云数据;根据点云数据构建头部的3D模型,以实现3D头部数据的采集;
S102.根据3D头部数据提取眼镜架的设计参数,形成眼镜架的3D模型;
S103.镜片模型:根据双眼视力状况和眼镜架的设计参数确定镜片参数,镜片参数包括:类型、度数、形状、材质和颜色,进而得到镜片的3D模型;
S104.效果显示:显示根据眼镜架模型和镜片模型合成的眼镜3D模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。即根据所述眼镜3D模型的显示效果重复步骤2和/或步骤3,选择并确定眼镜架的设计参数和镜片参数,完成眼镜设计。
本发明基于3D头部数据的眼镜自动设计方法,采用相机矩阵采集头部的特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特征的点云数据;根据所述点云数据构建头部的3D模型,以实现3D头部数据的采集;根据所述3D头部数据提取眼镜架的设计参数,形成眼镜架的3D模型;根据双眼视力状况和眼镜架的设计参数确定镜片参数,显示根据所述眼镜架模型和镜片模型合成的眼镜3D模型,根据显示效果调整眼镜架模型和/或镜片模型以完成眼镜设计。实现了眼镜的个性化、定制化、自动化设计和制造,设计制造的眼镜完全贴合佩戴者,提高了舒适度,采用相机矩阵控制技术进行 头部特征信息的采集,可以显著提高采集效率。
根据本发明一个实施例的基于3D头部数据的眼镜自动设计方法的模块流程:S201.用户登陆并输入性别、年龄、身高、体重等个人信息;S202.头部扫描设备采集用户的3D头部信息;S203.根据个人信息和3D头部信息构建用户头部的3D模型;S204.根据用户头部的3D模型分析提取适合该用户的眼镜架数据;S205.测量用户的双眼,获得双眼视力状况,根据视力状况获得镜片参数,主要包括镜片的类型、度数、形状、材质和颜色等参数,进而得到镜片的3D模型;S206.将镜片的3D模型输入到标准镜片数据中进行匹配,获得最合适的镜片;S207.综合汇总提取的眼镜架数据、标准眼镜架数据库中的眼镜架数据以及最合适的镜片数据,以满足用户的个性化需求;S208.得到定制化的眼镜3D数据模型;S209.还可以将上一步骤中获得的眼镜3D模型再次与用户的头部模型相比较,对眼镜3D模型再次调整并最终确定适合该用户的眼镜3D数据;S210.根据最终确定的眼镜3D数据,采用多种形式包括以3D打印的形式或者生成眼镜的生产数据进行装配式生产,获得定制化、个性化、舒适的眼镜,也可以通过视频影像等形式进行虚拟佩戴体验。
优选的,在本发明的一个实施例中,在完成眼镜设计后通过3D打印来实现眼镜的装配式生产,这样可以快速、低成本、高匹配的生产出适合用户头部、面部特征的眼镜,提升了用户体验。优选的,所述眼镜可以是无框眼镜、有框眼镜或半框眼镜,相比有框眼镜,无框眼镜主要包括:镜圈、镜腿、鼻梁和鼻托,其中镜圈、镜腿、鼻梁这三个部件最影响眼镜的佩戴舒适度,为眼镜设计和生产中最重要的参数;无框眼镜的镜圈、镜腿、鼻梁可以采用18K金材料,利用18K金材料通过3D打印机生产所述镜圈、镜腿和鼻梁。18K金材料可以回收再利用,因此用户可以在眼镜款式过时或不喜欢的时候进行以旧换新,无需增加太多成本就可以随意更换眼镜,进一步提升了使用体验。对于半框或全框等有框眼镜而言,均包括:镜圈、桩头、镜腿、鼻梁和鼻托,其中,可以通过3D打印机生产镜圈、镜腿、桩头和鼻梁;优选的,因为是高端定制化设计制造,镜圈、镜腿和桩头可采用性能较好的钛合金材料,3D打印完成后可以对这些部件进行打磨和镀金,实现个性化款式和颜色的设计。
优选的,在步骤S204中,根据用户头部的3D模型分析提取适合该用户的眼镜架数据,包括如下步骤:根据3D头部数据计算眼镜架作用区域的数据;以及根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数。上述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧 倾角和鼻梁的数据以及眼部眼眶和瞳距的数据。眼镜架作用区域是指:眼镜通过鼻托放在鼻梁上,镜腿挂在耳朵上,眼镜(眼镜架)通过鼻梁与镜腿直接作用于头部,具体体现在眼睛、鼻子、耳朵及周围。虽然眼镜架不与眼部直接接触,但眼镜定制时必须考虑瞳距和瞳高,以双眼瞳距为例,其主要用作决定眼镜片(镜框)的宽度。瞳高即瞳孔的中心高度,指眼睛瞳孔到眼镜下边缘最低点的垂直距离,其主要用于决定眼镜片(镜框)的高度。鼻子的信息主要影响鼻梁及鼻托的定制,眼镜通过鼻梁或者鼻托作用到鼻子上,鼻梁的宽度一般设计为与双眼在一条直线上鼻子的宽度,鼻托位于鼻子两侧,并与鼻子贴合。耳朵的信息主要表现在耳朵最高点到眼睛的垂直距离以及与眼睛的高度差这两个方面,耳朵最高点到眼睛垂直距离用于设计镜腿长度,耳朵最高点到眼睛的高度差可以为佩戴眼镜时镜腿的倾斜程度提供依据。另外,由于眼镜自身的重量,在佩戴的过程中会因重力缘故自动下降2-4mm,所以对于眼镜的设计而言,考虑眼镜的重量因素影响是实现定制眼镜的重要因素。本发明实施例中,考虑了上述眼镜架作用区域的数据及其影响,有针对性的设计眼镜架的参数,包括:镜腿、镜圈、镜框、鼻梁、鼻托和桩头的形状、尺寸、材质等参数。
优选的,步骤S204中根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数可以包括:
S21.建立立体三维坐标系:根据3D头部数据,基于左右眼外角点31和左右耳屏点32创建平面作为头部数据坐标系的法兰克福面30;
S22.镜腿轮廓线的提取,把眼镜腿抽象为一条曲线,作为镜腿参考面40,所述镜腿参考面为两眼镜腿的镜脚弯点至镜腿的螺栓中心部分所在的与法兰克福平行的平面;
S23.根据镜腿参考面所在的轮廓确定镜腿弯点长度和镜腿的轮廓。
优选的,步骤S204还包括:耳根外轮廓的提取:由于镜腿除了作用于人体头型面部外,其耳曲形态对耳部舒适性及整个眼镜在头部的平衡与稳定有影响,本发明通过采集耳朵根部轮廓数据为耳曲设计提供设计依据;眼鼻区域的脸部形态提取:眼镜框与鼻托作用于脸部的眼睛区域与鼻子区域,然而人体的脸部形态为不规则曲面,对眼镜与鼻子区域进行参数提取有很大难度,因此在3D头部模型中选择标记点,并获取标记点的3D坐标值对脸部参数进行提取,对标记点的3D坐标值进行主成分分析,根据标记点,计算眼睛区域、眼框区域和鼻子区域,得到鼻子两侧倾角,鼻托,鼻梁参数值;瞳距的提取,镜框设计时,通过眼睛的中心点距离及镜框宽度和镜面角获取瞳距。
优选的,在步骤S207中,还包括:根据标准眼镜架数据库中的眼镜架数据确定眼镜架的款式,将上述眼镜架的设计参数与所述款式的眼镜架进行尺寸匹配计算,根据计算结果形成眼镜架的3D模型。
进一步的,在步骤S207中,可以通过3D扫描仪扫描现有眼镜,获得眼镜架的款式数据,例如可以通过多台可见光相机组成的相机矩阵采集市面上时尚流行畅销的眼镜的3D数据,将这些数据保存至标准眼镜架数据库中供用户选择,尤其是对造型和款式的选择,并且根据由用户3D头部数据获得的眼镜架的设计参数,调整用户在标准眼镜架数据库中选择的眼镜框架款式对应的数据,即通过3D扫描仪逆向工程,得到匹配该用户的最佳数据,进而设计生产出用户满意的眼镜。
优选的,根据本发明的一个实施例,在对3D头部数据采集的步骤S101中,还包括如下步骤:
根据特征点的特征进行特征点的匹配,建立匹配的特征点数据集;
根据相机矩阵的光学信息,计算各台相机相对于头部在空间上的相对位置,并根据相对位置计算出多幅图像中的特征点的空间深度信息;
根据匹配的特征点数据集和特征点的空间深度信息,生成头部特征的点云数据。
优选的,多幅头部图像中各自的特征点的特征采用尺度不变特征转换SIFT特征描述子来描述。
优选的,根据相机矩阵的光学信息,计算各台相机相对于头部在空间上的相对位置的步骤进一步包括:
根据相机矩阵的光学信息,采用光束平差法计算各台相机相对于头部在空间上的相对位置。
优选的,多幅头部图像中的特征点的空间深度信息包括:空间位置信息和颜色信息。
优选的,根据点云数据构建头部的3D模型的步骤进一步包括:
设定待构建的3D模型的参考尺寸;
根据参考尺寸和特征点云数据的空间位置信息,确定特征点云数据中各个特征点的空间尺寸,从而构建3D头部模型。
步骤S101采用相机矩阵控制技术进行头部信息的采集,可以显著提高头部信息的采集效率;并且,本发明实施例利用采集到头部在空间上的特征信息,完整地复原头部在空间上的各项特征,为后续的头部数据的应用提供了无限的 可能性。
在本发明的可选实施例中,上文步骤S101中基于提取的多幅头部图像中各自的特征点,生成头部的特征点云数据,具体可以是包括以下步骤S1061至步骤S1063。上文已详细说明,在此不再赘述。
在此处所提供的说明书中,说明了大量具体细节。然而,能够理解,本发明的实施例可以在没有这些具体细节的情况下实践。在一些实例中,并未详细示出公知的方法、结构和技术,以便不模糊对本说明书的理解。
类似地,应当理解,为了精简本公开并帮助理解各个发明方面中的一个或多个,在上面对本发明的示例性实施例的描述中,本发明的各个特征有时被一起分组到单个实施例、图、或者对其的描述中。然而,并不应将该公开的方法解释成反映如下意图:即所要求保护的本发明要求比在每个权利要求中所明确记载的特征更多的特征。更确切地说,如下面的权利要求书所反映的那样,发明方面在于少于前面公开的单个实施例的所有特征。因此,遵循具体实施方式的权利要求书由此明确地并入该具体实施方式,其中每个权利要求本身都作为本发明的单独实施例。
本领域那些技术人员可以理解,可以对实施例中的设备中的模块进行自适应性地改变并且把它们设置在与该实施例不同的一个或多个设备中。可以把实施例中的模块或单元或组件组合成一个模块或单元或组件,以及此外可以把它们分成多个子模块或子单元或子组件。除了这样的特征和/或过程或者单元中的至少一些是相互排斥之外,可以采用任何组合对本说明书(包括伴随的权利要求、摘要和附图)中公开的所有特征以及如此公开的任何方法或者设备的所有过程或单元进行组合。除非另外明确陈述,本说明书(包括伴随的权利要求、摘要和附图)中公开的每个特征可以由提供相同、等同或相似目的的替代特征来代替。
此外,本领域的技术人员能够理解,尽管在此所述的一些实施例包括其它实施例中所包括的某些特征而不是其它特征,但是不同实施例的特征的组合意味着处于本发明的范围之内并且形成不同的实施例。例如,在权利要求书中,所要求保护的实施例的任意之一都可以以任意的组合方式来使用。
本发明的各个部件实施例可以以硬件实现,或者以在一个或者多个处理器上运行的软件模块实现,或者以它们的组合实现。本领域的技术人员应当理解,可以在实践中使用微处理器或者数字信号处理器(DSP)来实现根据本发明实施例的一些或者全部部件的一些或者全部功能。本发明还可以实现为用于执行 这里所描述的方法的一部分或者全部的设备或者装置程序(例如,计算机程序和计算机程序产品)。这样的实现本发明的程序可以存储在计算机可读介质上,或者可以具有一个或者多个信号的形式。这样的信号可以从因特网网站上下载得到,或者在载体信号上提供,或者以任何其他形式提供。
应该注意的是上述实施例对本发明进行说明而不是对本发明进行限制,并且本领域技术人员在不脱离所附权利要求的范围的情况下可设计出替换实施例。在权利要求中,不应将位于括号之间的任何参考符号构造成对权利要求的限制。单词“包含”不排除存在未列在权利要求中的元件或步骤。位于元件之前的单词“一”或“一个”不排除存在多个这样的元件。本发明可以借助于包括有若干不同元件的硬件以及借助于适当编程的计算机来实现。在列举了若干装置的单元权利要求中,这些装置中的若干个可以是通过同一个硬件项来具体体现。单词第一、第二、以及第三等的使用不表示任何顺序。可将这些单词解释为名称。
至此,本领域技术人员应认识到,虽然本文已详尽示出和描述了本发明的多个示例性实施例,但是,在不脱离本发明精神和范围的情况下,仍可根据本发明公开的内容直接确定或推导出符合本发明原理的许多其他变型或修改。因此,本发明的范围应被理解和认定为覆盖了所有这些其他变型或修改。
Claims (31)
- 一种基于3D头部数据的眼镜自动设计***,其特征在于,包括:图像采集模块,用于采集头部的多幅图像,所述图像采集模块包括由多台相机组成的相机矩阵;3D模型生成模块,用于提取所述多幅图像中的特征点,生成头部特征的点云数据,并根据所述点云数据生成头部的3D模型,完成3D头部数据的采集;眼镜架设计模块,用于根据所述3D头部数据提取眼镜架的设计参数进而形成眼镜架模型,具体为:根据3D头部数据计算眼镜架作用区域的数据;以及根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数;上述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据;提取出所述眼镜架的设计参数还包括:耳根外轮廓的提取:通过采集耳朵根部轮廓数据为耳曲设计提供设计依据;眼鼻区域的脸部形态提取:在3D头部模型中选择标记点,并获取标记点的3D坐标值对脸部参数进行提取,对标记点的3D坐标值进行主成分分析,根据标记点,计算眼睛区域、眼框区域和鼻子区域,得到鼻子两侧倾角,鼻托,鼻梁参数值;瞳距的提取:镜框设计时,通过眼睛的中心点距离及镜框宽度和镜面角获取瞳距;提取眼镜架的设计参数过程具体包括如下步骤:①建立三维坐标系:根据3D头部数据,基于左右眼外角点和左右耳屏点创建平面作为头部数据坐标系的法兰克福面;②镜腿轮廓线的提取:把眼镜腿抽象为一条曲线,作为镜腿参考面,所述镜腿参考面为两眼镜腿的镜脚弯点至镜腿的螺栓中心部分所在的与法兰克福平行的平面;③根据镜腿参考面所在的轮廓确定镜腿弯点长度和镜腿的轮廓;眼镜架设计模块通过多台可见光相机组成的相机矩阵采集眼镜的3D数据,将这些数据保存至标准眼镜架数据库中供用户选择,并且根据由用户3D头部数据获得的眼镜架的设计参数,调整用户在标准眼镜架数据库中选择的眼镜框架款式对应的数据;镜片设计模块,用于根据双眼视力状况和所述眼镜架的设计参数确定镜片参数,进而得到镜片模型;合成设计模块,用于根据所述眼镜架模型和镜片模型合成的眼镜3D模型;显示调整模块,用于显示所述眼镜3D模型的佩戴效果,并根据佩戴效果调整眼镜架模型和镜片模型以完成眼镜的设计。
- 根据权利要求1所述的眼镜自动设计***,其特征在于,所述3D模型生成模块包括:特征点提取单元,用于对所述多幅图像进行处理,提取所述多幅图像中各自的特征点;点云生成单元,用于根据所述特征点的特征进行特征点的匹配,建立匹配的特征点数据集,根据多台相机的光学信息,计算各台相机相对于头部在空间上的相对位置,根据所述相对位置计算出所述多幅图像中的特征点的空间深度信息,所述点云生产单元根据匹配的特征点数据集和特征点的空间深度信息,生成头部特征的点云数据;3D模型构建单元,用于根据所述头部特征的点云数据构建头部的3D模型,完成3D头部数据的采集。
- 根据权利要求2所述的眼镜自动设计***,其特征在于:所述特征点提取单元采用尺度不变特征转换SIFT特征描述子来描述所述多幅图像中各自特征点的特征;所述点云生成单元采用光束平差法计算各台相机相对于头部在空间上的相对位置;所述空间深度信息包括:空间位置信息和颜色信息。
- 根据权利要求2所述的眼镜自动设计***,其特征在于:所述3D模型构建单元用于设定待构建的3D模型的参考尺寸,并根据所述参考尺寸和空间位置信息确定所述点云数据中各个特征点的空间尺寸,从而构建3D头部模型。
- 根据权利要求1所述的眼镜自动设计***,其特征在于,所述眼镜架设计模块用于根据所述3D头部数据计算眼镜架作用区域的数据并根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数。
- 根据权利要求5所述的眼镜自动设计***,其特征在于:所述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据;所述眼镜架的设计参数包括:镜腿、镜圈、镜框、鼻梁、鼻托和桩头。
- 根据权利要求6所述的眼镜自动设计***,其特征在于,所述眼镜架设计模块还用于确定眼镜架的款式,眼镜架设计模块将所述眼镜架的设计参数与 所述款式的眼镜架的数据进行尺寸匹配计算,根据计算结果形成眼镜架模型。
- 根据权利要求7所述的眼镜自动设计***,其特征在于,所述眼镜架设计模块通过3D扫描仪扫描现有眼镜获得不同款式眼镜架的数据。
- 根据权利要求7所述的眼镜自动设计***,其特征在于,所述眼镜架设计模块用于根据由所述3D头部数据获得的眼镜架的设计参数对所述款式眼镜架的数据进行调整,进而形成眼镜架模型。
- 根据权利要求1所述的眼镜自动设计***,其特征在于,还包括3D打印设备,所述3D打印设备用于根据眼镜3D模型打印完成设计后的眼镜。
- 根据权利要求10所述的眼镜自动设计***,其特征在于,所述3D打印设备打印的眼镜为无框眼镜,所述无框眼镜包括:镜圈、镜腿、鼻梁和鼻托;所述无框眼镜的材质为18K金。
- 根据权利要求10所述的眼镜自动设计***,其特征在于,所述3D打印设备打印的眼镜为半框眼镜或有框眼镜,所述半框眼镜或有框眼镜均包括:镜圈、桩头、镜腿、鼻梁和鼻托;所述3D打印设备采用钛合金材料打印所述镜圈、镜腿、桩头和鼻梁。
- 根据权利要求1所述的眼镜自动设计***,其特征在于,所述图像采集模块还包括:支撑结构以及设置在所述支撑结构上的弧形承载结构,所述多台相机布置在所述弧形承载结构上形成所述相机矩阵;与所述支撑结构连接的底座,在所述底座上设有用于固定人体拍照位置的座椅,当人***于所述座椅上时,自动利用所述相机矩阵采集人体的头部信息。
- 根据权利要求13所述的眼镜自动设计***,其特征在于,所述弧形承载结构上还设有用于显示3D头部数据的显示器;所述显示器还用于设定各台相机的拍照参数。
- 一种基于3D头部数据的眼镜自动设计方法,其特征在于,包括步骤:步骤1.采用相机矩阵采集头部的特征信息,得到头部的多幅图像,提取所述多幅图像中的特征点,生成头部特征的点云数据;根据所述点云数据构建头部的3D模型,以实现3D头部数据的采集;步骤2.根据所述3D头部数据提取眼镜架的设计参数,形成眼镜架的3D模型,具体为:根据3D头部数据计算眼镜架作用区域的数据;以及根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数;上述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据;提取出所述眼镜架的设计参数还包括:耳根外轮廓的提取:通过采集耳朵根部轮廓数据为耳曲设计提供设计依据;眼鼻区域的脸部形态提取:在3D头部模型中选择标记点,并获取标记点的3D坐标值对脸部参数进行提取,对标记点的3D坐标值进行主成分分析,根据标记点,计算眼睛区域、眼框区域和鼻子区域,得到鼻子两侧倾角,鼻托,鼻梁参数值;瞳距的提取:镜框设计时,通过眼睛的中心点距离及镜框宽度和镜面角获取瞳距;提取眼镜架的设计参数过程具体包括如下步骤:①建立三维坐标系:根据3D头部数据,基于左右眼外角点和左右耳屏点创建平面作为头部数据坐标系的法兰克福面;②镜腿轮廓线的提取:把眼镜腿抽象为一条曲线,作为镜腿参考面,所述镜腿参考面为两眼镜腿的镜脚弯点至镜腿的螺栓中心部分所在的与法兰克福平行的平面;③根据镜腿参考面所在的轮廓确定镜腿弯点长度和镜腿的轮廓;眼镜架设计模块通过多台可见光相机组成的相机矩阵采集眼镜的3D数据,将这些数据保存至标准眼镜架数据库中供用户选择,并且根据由用户3D头部数据获得的眼镜架的设计参数,调整用户在标准眼镜架数据库中选择的眼镜框架款式对应的数据;步骤3.镜片模型:根据双眼视力状况和眼镜架的设计参数确定镜片参数,镜片参数包括:类型、度数、形状、材质和颜色,进而得到镜片的3D模型;步骤4.效果显示:显示根据所述眼镜架模型和镜片模型合成的眼镜3D模型,根据显示效果调整眼镜架模型和镜片模型以完成眼镜设计。
- 根据权利要求15所述的眼镜自动设计方法,其特征在于,步骤1进一步包括:根据所述特征点的特征进行特征点的匹配,建立匹配的特征点数据集;根据相机矩阵的光学信息,计算各台相机相对于头部在空间上的相对位置,并根据所述相对位置计算出所述多幅图像中的特征点的空间深度信息;根据匹配的特征点数据集和特征点的空间深度信息,生成头部特征的点云数据。
- 根据权利要求16所述的眼镜自动设计方法,其特征在于:所述多幅头部图像中各自的特征点的特征采用尺度不变特征转换SIFT特征描述子来描述。
- 根据权利要求16所述的眼镜自动设计方法,其特征在于:所述根据相机矩阵的光学信息,计算各台相机相对于头部在空间上的相对位置的步骤进一步包括:根据相机矩阵的光学信息,采用光束平差法计算各台相机相对于头部在空间上的相对位置。
- 根据权利要求16所述的眼镜自动设计方法,其特征在于:所述多幅头部图像中的特征点的空间深度信息包括:空间位置信息和颜色信息。
- 根据权利要求16所述的眼镜自动设计方法,其特征在于:根据点云数据构建头部的3D模型的步骤进一步包括:设定待构建的3D模型的参考尺寸;根据所述参考尺寸和所述特征点云数据的空间位置信息,确定所述特征点云数据中各个特征点的空间尺寸,从而构建标定尺寸的3D头部模型。
- 根据权利要求15所述的眼镜自动设计方法,其特征在于,步骤1中:将所述多幅图像输入基于SIFT算法构建的配准算法模型,进行配准计算得到配准数据,基于所述配准数据生成3D头部数据。
- 根据权利要求15所述的眼镜自动设计方法,其特征在于,步骤2进一步包括:根据所述3D头部数据计算眼镜架作用区域的数据;以及根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数。
- 根据权利要求22所述的眼镜自动设计方法,其特征在于:所述眼镜架作用区域的数据包括:耳部左右耳屏点和耳根外轮廓的数据、鼻部鼻子两侧倾角和鼻梁的数据以及眼部眼眶和瞳距的数据;所述眼镜架的设计参数包括:镜腿、镜圈、镜框、鼻梁、鼻托和桩头。
- 根据权利要求23所述的眼镜自动设计方法,其特征在于,步骤2进一步包括:确定所述眼镜架的款式,将所述眼镜架的设计参数与所述款式的眼镜架进行尺寸匹配计算,根据计算结果形成眼镜架的3D模型。
- 根据权利要求24所述的眼镜自动设计方法,其特征在于,步骤2进一步包括:通过3D扫描仪扫描现有眼镜,获得所述眼镜架的款式数据。
- 根据权利要求25所述的眼镜自动设计方法,其特征在于,所述尺寸匹配计算包括:根据由所述3D头部数据获得的眼镜架的设计参数,对所述眼镜架的款式数据进行调整。
- 根据权利要求23所述的眼镜自动设计方法,其特征在于,根据所述眼镜架作用区域的数据提取出所述眼镜架的设计参数包括:S21.建立立体三维坐标系:根据3D头部数据,基于左右眼外角点和左右耳屏点创建平面作为头部数据坐标系的法兰克福面;S22.镜腿轮廓线的提取:把眼镜腿抽象为一条曲线,作为镜腿参考面,所述镜腿参考面为两眼镜腿的镜脚弯点至镜腿的螺栓中心部分所在的与法兰克福平行的平面;S23.根据镜腿参考面所在的轮廓设计镜腿弯点长度和镜腿的轮廓。
- 根据权利要求15所述的眼镜自动设计方法,其特征在于,步骤4进一步包括:根据所述眼镜3D模型的显示效果重复步骤2和/或步骤3,选择并确定眼镜架的设计参数和镜片参数,完成眼镜设计。
- 根据权利要求28所述的眼镜自动设计方法,其特征在于,步骤4进一步包括:完成眼镜设计后通过3D打印实现眼镜的装配式生产。
- 根据权利要求29所述的眼镜自动设计方法,其特征在于,所述眼镜为无框眼镜,所述无框眼镜包括:镜圈、镜腿、鼻梁和鼻托;所述无框眼镜的材质为18K金,通过3D打印机生产所述镜圈、镜腿和鼻梁。
- 根据权利要求29所述的眼镜自动设计方法,其特征在于,所述眼镜为半框眼镜或有框眼镜,所述半框眼镜或有框眼镜均包括:镜圈、桩头、镜腿、鼻梁和鼻托;通过3D打印机生产所述镜圈、镜腿、桩头和鼻梁;所述镜圈、镜腿和桩头采用钛合金材料。
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810152228.0A CN108490641B (zh) | 2018-02-14 | 2018-02-14 | 基于3d头部数据的眼镜自动设计*** |
CN201810152229.5 | 2018-02-14 | ||
CN201810152228.0 | 2018-02-14 | ||
CN201810152229.5A CN108490642A (zh) | 2018-02-14 | 2018-02-14 | 基于3d头部数据的眼镜自动设计方法 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2019157988A1 true WO2019157988A1 (zh) | 2019-08-22 |
Family
ID=67619747
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2019/074431 WO2019157988A1 (zh) | 2018-02-14 | 2019-02-01 | 基于3d头部数据的眼镜自动设计***及设计方法 |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2019157988A1 (zh) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102073759A (zh) * | 2010-12-29 | 2011-05-25 | 温州大学 | 一种基于脸型特征参数的眼镜配置控制方法 |
CN106910102A (zh) * | 2016-07-25 | 2017-06-30 | 湖南拓视觉信息技术有限公司 | 眼镜虚拟试戴方法及装置 |
CN107085864A (zh) * | 2017-06-01 | 2017-08-22 | 北京大学第三医院 | 基于特征点的眼镜建模装置及方法、眼镜制作方法及眼镜 |
CN107103513A (zh) * | 2017-04-23 | 2017-08-29 | 广州帕克西软件开发有限公司 | 一种眼镜虚拟试戴方法 |
CN108490641A (zh) * | 2018-02-14 | 2018-09-04 | 天目爱视(北京)科技有限公司 | 基于3d头部数据的眼镜自动设计*** |
CN108490642A (zh) * | 2018-02-14 | 2018-09-04 | 天目爱视(北京)科技有限公司 | 基于3d头部数据的眼镜自动设计方法 |
-
2019
- 2019-02-01 WO PCT/CN2019/074431 patent/WO2019157988A1/zh active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102073759A (zh) * | 2010-12-29 | 2011-05-25 | 温州大学 | 一种基于脸型特征参数的眼镜配置控制方法 |
CN106910102A (zh) * | 2016-07-25 | 2017-06-30 | 湖南拓视觉信息技术有限公司 | 眼镜虚拟试戴方法及装置 |
CN107103513A (zh) * | 2017-04-23 | 2017-08-29 | 广州帕克西软件开发有限公司 | 一种眼镜虚拟试戴方法 |
CN107085864A (zh) * | 2017-06-01 | 2017-08-22 | 北京大学第三医院 | 基于特征点的眼镜建模装置及方法、眼镜制作方法及眼镜 |
CN108490641A (zh) * | 2018-02-14 | 2018-09-04 | 天目爱视(北京)科技有限公司 | 基于3d头部数据的眼镜自动设计*** |
CN108490642A (zh) * | 2018-02-14 | 2018-09-04 | 天目爱视(北京)科技有限公司 | 基于3d头部数据的眼镜自动设计方法 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108490641B (zh) | 基于3d头部数据的眼镜自动设计*** | |
US11592691B2 (en) | Systems and methods for generating instructions for adjusting stock eyewear frames using a 3D scan of facial features | |
US10983368B2 (en) | Systems and methods for creating eyewear with multi-focal lenses | |
US11428958B2 (en) | Method and system to create custom, user-specific eyewear | |
JP3250184B2 (ja) | 眼鏡装用シミュレーションシステム | |
CN108490642A (zh) | 基于3d头部数据的眼镜自动设计方法 | |
CN103690171A (zh) | 眼镜测配装置、眼镜测配服务端和眼镜测配方法 | |
WO2019157988A1 (zh) | 基于3d头部数据的眼镜自动设计***及设计方法 | |
CN110866970B (zh) | 一种通过面部关键点识别实现重建配镜的***及方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 19754826 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 19754826 Country of ref document: EP Kind code of ref document: A1 |