CN109741386A - The Enhancement Method and stereo visual system of stereo visual system - Google Patents
The Enhancement Method and stereo visual system of stereo visual system Download PDFInfo
- Publication number
- CN109741386A CN109741386A CN201811603324.9A CN201811603324A CN109741386A CN 109741386 A CN109741386 A CN 109741386A CN 201811603324 A CN201811603324 A CN 201811603324A CN 109741386 A CN109741386 A CN 109741386A
- Authority
- CN
- China
- Prior art keywords
- infrared
- infrared image
- image
- visual system
- stereo
- 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.)
- Granted
Links
Abstract
The present invention provides a kind of Enhancement Method of stereo visual system and stereo visual system, multiple first infrared images by active illumination of stereo visual system shooting include the hot spot of array distribution in first infrared image；Then third infrared image is successively obtained greater than the edge or texture of spot size less than or equal to the edge or texture and resizing of spot size using low-pass filtering and high-pass filtering removal size, to in the case where not increasing any additional hardware, it obtains the infrared image for not including hot spot texture and does not increase new block a little, using the third infrared image and multiple available steady depth images of first infrared image, to improve the robustness of stereo visual system.
Description
Technical field
The present invention relates to the Enhancement Methods of visual pattern processing technology field more particularly to a kind of stereo visual system and vertical
Body vision system.
Background technique
Binocular stereo vision (Binocular Stereo Vision) is a kind of important form of machine vision, it is base
In principle of parallax and two images of the imaging device from different position acquisition visual scenes are utilized, by calculating image corresponding points
Between position deviation, the method to obtain object dimensional geological information.Binocular stereo vision merges the original that two " eyes " obtain
Beginning image simultaneously observes the difference between them, can obtain apparent sense of depth, establish the corresponding relationship between feature, by same sky
Between photosites of the physical points in different images be mapped, pass through the available parallax of this difference (Disparity) image.
Binocular stereo vision can distinguish active stereo vision (actively providing light source) according to the difference of light source and passively stand
Body vision (by environment light), the light source (dot matrix laser infrared source) that wherein active stereo vision passes through increase pseudo random pattern
Scene is textured, to make up passive stereo vision in the deficiency for repeating texture or weak texture scene, pseudo random pattern
The enhanced X-ray source texture of scene, reduce the complexity of Stereo Matching Algorithm, improve the precision of depth image.
In order to enhance the matched robustness of active stereo, a pair of or a width " clean " infrared image is usually also needed, i.e.,
Infrared image not comprising the hot spot texture that the illumination of pseudo random pattern light source generates, the existing infrared image for obtaining " clean "
Method is usually to increase a visible image capturing head and notch filter or one infrared camera of increase and an infrared floodlight
Lamp, but on the one hand both methods increases system cost, on the one hand increase system dimension, it can also be because of three cameras
Parallax reason, cause in scene more information to be blocked (not visible simultaneously by three cameras), so that it cannot by standing
Body matching algorithm is correctly handled, the depth map inaccuracy caused.
Summary of the invention
The purpose of the present invention is to provide a kind of Enhancement Method of stereo visual system and stereo visual systems, are not increasing
In the case where any additional hardware, the infrared figure of " clean " can be obtained, and not will increase new block a little.
In order to achieve the above object, the present invention provides a kind of Enhancement Methods of stereo visual system, comprising:
Obtain multiple first infrared images by active illumination of stereo visual system shooting, first infrared image
In include array distribution hot spot；
It is small to remove size in first infrared image that low-pass filtering is carried out at least one described first infrared image
In or equal to spot size edge or texture and formed the second infrared image；
High-pass filtering is carried out to second infrared image to restore size in second infrared image and be greater than hot spot ruler
Very little edge or texture simultaneously form third infrared image；
Steady depth image is obtained according to the third infrared image and multiple first infrared images.
Optionally, the low-pass filtering includes weighted mean filter, and the high-pass filtering includes differential filtering.
Optionally, the step of steady depth image is obtained according to the third infrared image and multiple first infrared images
Suddenly include:
Utilize the cost function of multiple first infrared image building Stereo matchings；
Utilize the regularization term of the cost function of third infrared image building Stereo matching；
By the cost function of Stereo matching and its energy function of regularization term common definition Stereo matching, pass through solution
The minimum value of the energy function of each pixel obtains the parallax value of each pixel, to obtain original depth image.
Optionally, after obtaining the original depth-map, the Enhancement Method of the stereo visual system further include:
It carries out guarantor side as weight figure using the infrared figure of the third to filter, wherein the guarantor side filtering includes that joint adds
Weigh any one of mean filter, bilateral filtering, Steerable filter or Weighted median filtering.
Optionally, after obtaining the steady depth image, the Enhancement Method of the stereo visual system further include:
Threedimensional model is established according to the steady depth image and using the third infrared image as texture maps, in institute
It states and sticks texture on threedimensional model.
The present invention provides a kind of stereo visual systems, comprising:
Infrared array light source module, for issuing infrared dot matrix beam lighting to illuminate to visual scene；
Multiple infrared photography modules, for shooting by multiple first infrared images of active illumination；
Low-pass filtering module, for carrying out low-pass filtering at least one described first infrared image to remove described first
Edge or texture and formation second infrared image of the size less than or equal to spot size in infrared image；
High-pass filtering module, for carrying out high-pass filtering to second infrared image to restore second infrared image
Middle size is greater than the edge of spot size or texture and forms third infrared image；
Computing module, for obtaining steady depth map according to the third infrared image and multiple first infrared images
Picture.
Optionally, the low-pass filtering includes weighted mean filter, and the high-pass filtering includes differential filtering.
Optionally, the computing module includes:
Stereo matching module, the stereo matching module utilize multiple first infrared images and the infrared figure of the third
Picture constructs energy function and solves the minimum value of the energy function of each pixel, and according to the energy function of each pixel
Minimum value obtains the parallax value of each pixel to obtain original depth image；
Post-processing module, the post-processing module is using the third infrared image as weight figure, to the original depth
Degree image carries out protecting side filtering, to obtain the steady depth image.
Optionally, the stereo visual system further includes three-dimensional reconstruction module, and the three-dimensional reconstruction module is according to described steady
Strong depth image establishes threedimensional model and using the third infrared image as texture maps, sticks line on the threedimensional model
Reason.
Optionally, the stereo visual system is binocular active stereo vision system.
In the Enhancement Method of stereo visual system provided by the invention and stereo visual system, stereo visual system shooting
Multiple first infrared images by active illumination, include the hot spot of array distribution in first infrared image；Then successively
Simultaneously resizing is greater than light at the edge or texture for being less than or equal to spot size using low-pass filtering and high-pass filtering removal size
The edge or texture of spot size obtain third infrared image, to not wrapped in the case where not increasing any additional hardware
The infrared image of the texture containing hot spot and do not increase it is new block a little, using the third infrared image and multiple described first infrared
The available steady depth image of image, to improve the robustness of stereo visual system.
Detailed description of the invention
Fig. 1 is the flow chart of the Enhancement Method of stereo visual system provided in an embodiment of the present invention；
Fig. 2 is the another flow chart of the Enhancement Method of stereo visual system provided in an embodiment of the present invention；
Fig. 3 is the structural block diagram of stereo visual system provided in an embodiment of the present invention；
Wherein, appended drawing reference are as follows:
The infrared array light source module of 1-；2- infrared photography module；3- low-pass filtering module；4- high-pass filtering module；5- meter
Calculate module.
Specific embodiment
A specific embodiment of the invention is described in more detail below in conjunction with schematic diagram.According to following description and
Claims, advantages and features of the invention will become apparent from.It should be noted that attached drawing is all made of very simplified form and
Using non-accurate ratio, only for the purpose of facilitating and clarifying the purpose of the embodiments of the invention.
As shown in Figure 1, its flow chart for the Enhancement Method of stereo visual system provided in this embodiment, the stereopsis
The Enhancement Method of feel system includes:
S1: multiple first infrared images by active illumination of stereo visual system shooting are obtained, described first is infrared
It include the hot spot of array distribution in image；
S2: low-pass filtering is carried out to remove size in first infrared image at least one described first infrared image
Less than or equal to the edge or texture and the second infrared image of formation of spot size；
S3: high-pass filtering is carried out to second infrared image to restore size in second infrared image and be greater than hot spot
The edge or texture of size simultaneously form third infrared image；
S4: steady depth image is obtained according to the third infrared image and multiple first infrared images.
Specifically, the stereo visual system is active stereo visual system, such as there are two red for tool in the present embodiment
Outer photographing module shoots infrared image to imitate left eye and the right eye of human body, and the active stereo vision system has infrared
Array light source module, the infrared array light source module is, for example, dot matrix laser, can issue dot matrix infrared beam with right
One visual scene carries out active illumination.
As shown in Fig. 2, starting to execute step S1 after carrying out active illumination to the visual scene, the stereoscopic vision is obtained
Multiple first infrared images of system photographs include the hot spot of array distribution in first infrared image.Specifically, described vertical
Two infrared photography modules of body vision system shoot same visual scene respectively to obtain two the first infrared images (including left side
First infrared image and right first infrared image), since the visual scene is by active illumination, so each described first is infrared
Hot spot (profile of dot matrix hot spot) is all had in image.
Then step S2 is executed, at least one described first infrared image is used to carry out low-pass filtering to remove described first
Edge or texture and formation second infrared image of the size less than or equal to spot size in infrared image.Specifically, this implementation
Tool is there are two first infrared image in example, can be first red to any one first infrared image or two according to actual needs
Outer image carries out low-pass filtering, to remove the edge or texture that size in first infrared image is less than or equal to spot size
To form second infrared image (including left second infrared image and right second infrared image).It is described low in the present embodiment
Pass filter is weighted mean filter, and the dot matrix infrared beam that the size of filtering can be issued according to infrared array light source module carries out
Adjustment, as long as size in first infrared image can be less than or equal to the edge or line of spot size after realizing low-pass filtering
Reason removal.
Further, after forming second infrared image to first infrared image progress low-pass filtering, described first
High frequency section in infrared image may lose during low-pass filtering.Next step S3 is executed, it is red to described second
Outer image carries out high-pass filtering to restore edge or texture of the size greater than spot size in second infrared image and be formed
Third infrared image.It is understood that the high frequency section in first infrared image shows as the edge or texture of high frequency,
After carrying out high-pass filtering to second infrared image, the edge or texture that can restore the second infrared image medium-high frequency are simultaneously
Form the third infrared image (including left third infrared image and right third infrared image).In the present embodiment, the high pass
It is filtered into differential filtering.
Further, the third infrared image formed after the low-pass filtering and high-pass filtering is the infrared figure of " clean "
Picture can improve the robustness of the stereo visual system using the third infrared image.
It is possible, firstly, to using the cost function of multiple first infrared image building Stereo matchings, followed by described
Third infrared image constructs the regularization term of the cost function of Stereo matching, and Stereo Matching Algorithm is by seeking energy function most
Small value acquires best match, can be obtained the parallax of each pixel according to the minimum value unpacking of the energy function acquired
Value, then obtains the corresponding depth information of each pixel according to the parallax value of each pixel, so as to obtain depth map
Picture, so, in the present embodiment, pass through the cost function of Stereo matching and its energy letter of regularization term common definition Stereo matching
Number obtains the parallax value of each pixel by solving the minimum value of energy function of each pixel, to obtain original depth
Image.However the general type of energy function construction is that matching cost adds regularization term, wherein matching cost measurement is multiple
The similarity degree of the first infrared image respective pixel, regularization term by introduce penalty factor make the gradient of energy function with
Multiple first infrared images are consistent, if without the third infrared image but using any first infrared image
The constraint of regularization is done, then the depth information of each pixel acquired can include the gradient information of the hot spot, also to obtain the final product
It can see that light spot profile, the present embodiment are come using the gradient information of the third infrared image in the original depth image out
Bound energy function, so that obtained original depth image does not have light spot profile, and edge can be with the third infrared image pair
Together.
Secondly as consumption is very long when Stereo Matching Algorithm, the optimizing algorithm generally for energy function is all local
Or half global, original depth image counted so does not have the characteristic of global optimum.So obtaining original depth-map
As after, post-processing is filtered to the original depth image using the third infrared image, to remove local optimal searching mistake
The noise being introduced into the original depth image.For example, using the third infrared image as weight figure, to the original depth
Degree image protect side filtering to form steady depth image, and the steady depth image has been filtered out noise well and protected
Edge is held to be aligned with the third infrared image.Which kind of the present invention is not limited to use the guarantor side of concrete form to filter, for example combine
It is any in weighted mean filter, bilateral filtering, Steerable filter or Weighted median filtering.Further, what is obtained is described
Steady depth image only includes that depth information does not include texture information, and for the ease of showing, the present embodiment also utilizes described steady
Depth image carries out three-dimensional reconstruction to form the threedimensional model of the corresponding visual scene, and the third infrared image is recycled to make
For texture maps, the texture of visual scene is sticked in the threedimensional model, since the third infrared image does not include the light
Speckle reason, so hot spot texture will not be sticked as texture maps in the threedimensional model using the third infrared image,
It will not influence the bandwagon effect of threedimensional model.
Based on this, this implementation additionally provides a kind of stereo visual system characterized by comprising
Infrared array light source module 1, for issuing infrared dot matrix beam lighting to illuminate to visual scene；
Multiple infrared photography modules 2, for shooting by multiple first infrared images of active illumination；
Low-pass filtering module 3 removes described for carrying out low-pass filtering at least one described first infrared image
Edge or texture and formation second infrared image of the size less than or equal to spot size in one infrared image；
High-pass filtering module 4, for carrying out high-pass filtering to second infrared image to restore the described second infrared figure
Size is greater than the edge of spot size or texture and forms third infrared image as in；
Computing module 5, for obtaining steady depth according to the third infrared image and multiple first infrared images
Image.
Specifically, the stereo visual system is binocular active stereo vision system, there are two infrared photography modules for tool
2.In the present embodiment, the low-pass filtering module 3 is Weighted Mean Filter, and the high-pass filtering module 4 is difference filter.
Optionally, the computing module includes stereo matching module and post-processing module, and the stereo matching module utilizes
Two first infrared images and obtained by any first infrared image that two infrared photography modules 2 are shot
Third infrared image building energy function and the energy function that solves each pixel minimum value, and according to each picture
The minimum value of the energy function of vegetarian refreshments obtains the parallax value of each pixel to obtain original depth image, the post-processing module
Using the third infrared image as guidance diagram, the original depth image is carried out to protect side filtering, it is described steady to obtain
Depth image.
Optionally, the stereo visual system further includes three-dimensional reconstruction module, and the three-dimensional reconstruction module is according to described steady
Strong depth image establishes threedimensional model and using the third infrared image as texture maps, sticks line on the threedimensional model
Reason, so that it further includes the texture information of visual scene that the threedimensional model, which not only includes the depth information of visual scene,.
To sum up, three-dimensional in the Enhancement Method of stereo visual system provided in an embodiment of the present invention and stereo visual system
Multiple first infrared images by active illumination of vision system shooting include the light of array distribution in first infrared image
Spot；Then edge or texture of the size less than or equal to spot size are successively removed using low-pass filtering and high-pass filtering and restored
Size is greater than the edge of spot size or texture obtains third infrared image, thus the case where not increasing any additional hardware
Under, obtain the infrared image for not including hot spot texture and do not increase it is new block a little, using the third infrared image and multiple
The available steady depth image of first infrared image, to improve the robustness of stereo visual system.
The above is only a preferred embodiment of the present invention, does not play the role of any restrictions to the present invention.Belonging to any
Those skilled in the art, in the range of not departing from technical solution of the present invention, to the invention discloses technical solution and
Technology contents make the variation such as any type of equivalent replacement or modification, belong to the content without departing from technical solution of the present invention, still
Within belonging to the scope of protection of the present invention.
Claims (10)
1. a kind of Enhancement Method of stereo visual system characterized by comprising
Multiple first infrared images by active illumination of stereo visual system shooting are obtained, are wrapped in first infrared image
Hot spot containing array distribution；
Low-pass filtering carried out at least one described first infrared image be less than with to remove size in first infrared image or
Equal to the edge or texture and the second infrared image of formation of spot size；
High-pass filtering is carried out to second infrared image to restore size in second infrared image and be greater than spot size
Edge or texture simultaneously form third infrared image；
Steady depth image is obtained according to the third infrared image and multiple first infrared images.
2. the Enhancement Method of stereo visual system as described in claim 1, which is characterized in that the low-pass filtering includes weighting
Mean filter, the high-pass filtering include differential filtering.
3. the Enhancement Method of stereo visual system as described in claim 1, which is characterized in that according to the third infrared image
And multiple first infrared images the step of obtaining steady depth image, includes:
Utilize the cost function of multiple first infrared image building Stereo matchings；
Utilize the regularization term of the cost function of third infrared image building Stereo matching；
It is each by solving by the cost function of Stereo matching and its energy function of regularization term common definition Stereo matching
The minimum value of the energy function of pixel obtains the parallax value of each pixel, to obtain original depth image.
4. the Enhancement Method of stereo visual system as claimed in claim 3, which is characterized in that obtain the original depth-map it
Afterwards, the Enhancement Method of the stereo visual system further include:
It carries out protecting side filtering as weight figure using the infrared figure of the third, wherein the guarantor side filtering includes combining weighting
Any one of value filtering, bilateral filtering, Steerable filter or Weighted median filtering.
5. the Enhancement Method of stereo visual system as described in claim 1, which is characterized in that obtain the steady depth image
Later, the Enhancement Method of the stereo visual system further include:
Threedimensional model is established according to the steady depth image and using the third infrared image as texture maps, described three
Texture is sticked on dimension module.
6. a kind of stereo visual system characterized by comprising
Infrared array light source module, for issuing infrared dot matrix beam lighting to illuminate to visual scene；
Multiple infrared photography modules, for shooting by multiple first infrared images of active illumination；
Low-pass filtering module, it is infrared to remove described first for carrying out low-pass filtering at least one described first infrared image
Edge or texture and formation second infrared image of the size less than or equal to spot size in image；
High-pass filtering module, for carrying out high-pass filtering to second infrared image to restore ruler in second infrared image
Very little edge or texture greater than spot size simultaneously forms third infrared image；
Computing module, for obtaining steady depth image according to the third infrared image and multiple first infrared images.
7. stereo visual system as claimed in claim 6, which is characterized in that the low-pass filtering includes weighted mean filter,
The high-pass filtering includes differential filtering.
8. stereo visual system as claimed in claim 6, which is characterized in that the computing module includes:
Stereo matching module, the stereo matching module utilize multiple first infrared images and the third infrared image structure
It builds energy function and solves the minimum value of the energy function of each pixel, and the minimum of the energy function according to each pixel
Value obtains the parallax value of each pixel to obtain original depth image；
Post-processing module, the post-processing module is using the third infrared image as weight figure, to the original depth-map
As carrying out protecting side filtering, to obtain the steady depth image.
9. stereo visual system as claimed in claim 8, which is characterized in that the stereo visual system further includes three-dimensional reconstruction
Module, the three-dimensional reconstruction module are established threedimensional model according to the steady depth image and are made using the third infrared image
For texture maps, texture is sticked on the threedimensional model.
10. the stereo visual system as described in any one of claim 6-9, which is characterized in that the stereo visual system is
Binocular active stereo vision system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811603324.9A CN109741386B (en) | 2018-12-26 | 2018-12-26 | Enhancement method of stereoscopic vision system and stereoscopic vision system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811603324.9A CN109741386B (en) | 2018-12-26 | 2018-12-26 | Enhancement method of stereoscopic vision system and stereoscopic vision system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109741386A true CN109741386A (en) | 2019-05-10 |
| CN109741386B CN109741386B (en) | 2020-07-31 |
Family
ID=66361338
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811603324.9A Active CN109741386B (en) | 2018-12-26 | 2018-12-26 | Enhancement method of stereoscopic vision system and stereoscopic vision system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109741386B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103839258A (en) * | 2014-02-13 | 2014-06-04 | 西安交通大学 | Depth perception method of binarized laser speckle images |
| CN104084669A (en) * | 2014-06-20 | 2014-10-08 | 河北工业大学 | Visual detecting method and device based on decoupling detection for circular seams |
| WO2014172228A1 (en) * | 2013-04-15 | 2014-10-23 | Microsoft Corporation | Robust stereo depth system |
| CN105004274A (en) * | 2015-07-07 | 2015-10-28 | 深圳大学 | Cylindrical surface radius measuring method based on three-dimensional vision |
| CN107392954A (en) * | 2017-07-04 | 2017-11-24 | 大连理工大学 | A kind of gross error point elimination method based on sequence image |
| CN107917701A (en) * | 2017-12-28 | 2018-04-17 | 人加智能机器人技术（北京）有限公司 | Measuring method and RGBD camera systems based on active binocular stereo vision |
| CN108389260A (en) * | 2018-03-19 | 2018-08-10 | 中国计量大学 | A kind of three-dimensional rebuilding method based on Kinect sensor |
| CN109087266A (en) * | 2018-08-09 | 2018-12-25 | 苏州大学 | A kind of image speckle iteration reduction method of combination bilateral filtering |
-
2018
- 2018-12-26 CN CN201811603324.9A patent/CN109741386B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014172228A1 (en) * | 2013-04-15 | 2014-10-23 | Microsoft Corporation | Robust stereo depth system |
| CN103839258A (en) * | 2014-02-13 | 2014-06-04 | 西安交通大学 | Depth perception method of binarized laser speckle images |
| CN104084669A (en) * | 2014-06-20 | 2014-10-08 | 河北工业大学 | Visual detecting method and device based on decoupling detection for circular seams |
| CN105004274A (en) * | 2015-07-07 | 2015-10-28 | 深圳大学 | Cylindrical surface radius measuring method based on three-dimensional vision |
| CN107392954A (en) * | 2017-07-04 | 2017-11-24 | 大连理工大学 | A kind of gross error point elimination method based on sequence image |
| CN107917701A (en) * | 2017-12-28 | 2018-04-17 | 人加智能机器人技术（北京）有限公司 | Measuring method and RGBD camera systems based on active binocular stereo vision |
| CN108389260A (en) * | 2018-03-19 | 2018-08-10 | 中国计量大学 | A kind of three-dimensional rebuilding method based on Kinect sensor |
| CN109087266A (en) * | 2018-08-09 | 2018-12-25 | 苏州大学 | A kind of image speckle iteration reduction method of combination bilateral filtering |
Non-Patent Citations (2)
| Title |
|---|
| YINDA ZHANG，ET AL.: "ActiveStereoNet: End-to-End Self-Supervised Learning for Active Stereo Systems", 《ARXIV》 * |
| 尹宝银，等: "图像上散斑的消除", 《咸阳师范学院学报》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109741386B (en) | 2020-07-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106355570B (en) | A kind of binocular stereo vision matching method of combination depth characteristic | |
| CN103279982B (en) | The speckle three-dimensional rebuilding method of the quick high depth resolution of robust | |
| CN102368137B (en) | Embedded calibrating stereoscopic vision system | |
| CN106469431A (en) | Image processing apparatus and photographic attachment | |
| CN110044300A (en) | Amphibious 3D vision detection device and detection method based on laser | |
| CN107560592A (en) | A kind of precision ranging method for optronic tracker linkage target | |
| CN105654547B (en) | Three-dimensional rebuilding method | |
| CN107274483A (en) | A kind of object dimensional model building method | |
| CN106447661A (en) | Rapid depth image generating method | |
| CN104539928A (en) | Three-dimensional printing image synthesizing method for optical grating | |
| CN103763543B (en) | The acquisition method of resultant hologram | |
| CN106355621A (en) | Method for acquiring depth information on basis of array images | |
| CN109285189A (en) | A kind of straight path quick calculation method synchronous without binocular | |
| CN106203429A (en) | Based on the shelter target detection method under binocular stereo vision complex background | |
| CN109242898A (en) | A kind of three-dimensional modeling method and system based on image sequence | |
| CN109461206A (en) | A kind of the face three-dimensional reconstruction apparatus and method of multi-view stereo vision | |
| CN106170086A (en) | The method of drawing three-dimensional image and device, system | |
| CN107801017A (en) | A kind of more mesh 3D video cameras and its 3D rendering modeling method | |
| CN114359406A (en) | Calibration of auto-focusing binocular camera, 3D vision and depth point cloud calculation method | |
| JP6285686B2 (en) | Parallax image generation device | |
| CN107864372A (en) | Solid picture-taking method, apparatus and terminal | |
| CN114511608A (en) | Method, device, terminal, imaging system and medium for acquiring depth image | |
| CN109029380B (en) | Stereo visual system and its calibration distance measuring method based on film coated type multispectral camera | |
| CN108090930A (en) | Barrier vision detection system and method based on binocular solid camera | |
| CN109741386A (en) | The Enhancement Method and stereo visual system of stereo visual system |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |