CN106023170A - Binocular 3D distortion rectification method based on GPU - Google Patents
Binocular 3D distortion rectification method based on GPU Download PDFInfo
- Publication number
- CN106023170A CN106023170A CN201610318480.5A CN201610318480A CN106023170A CN 106023170 A CN106023170 A CN 106023170A CN 201610318480 A CN201610318480 A CN 201610318480A CN 106023170 A CN106023170 A CN 106023170A
- Authority
- CN
- China
- Prior art keywords
- gpu
- left eyes
- picture
- robj
- lobj
- 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
- 238000000034 method Methods 0.000 title claims abstract description 22
- 238000013519 translation Methods 0.000 claims abstract description 8
- 238000012216 screening Methods 0.000 claims abstract description 6
- 238000012937 correction Methods 0.000 claims description 28
- 238000003709 image segmentation Methods 0.000 claims description 11
- 238000003384 imaging method Methods 0.000 claims description 9
- 239000000284 extract Substances 0.000 claims description 5
- FGUUSXIOTUKUDN-IBGZPJMESA-N C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 Chemical compound C1(=CC=CC=C1)N1C2=C(NC([C@H](C1)NC=1OC(=NN=1)C1=CC=CC=C1)=O)C=CC=C2 FGUUSXIOTUKUDN-IBGZPJMESA-N 0.000 claims description 3
- 238000000605 extraction Methods 0.000 claims description 3
- 238000011144 upstream manufacturing Methods 0.000 claims description 3
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 230000001105 regulatory effect Effects 0.000 abstract description 3
- 230000003247 decreasing effect Effects 0.000 abstract 1
- VYPSYNLAJGMNEJ-UHFFFAOYSA-N Silicium dioxide Chemical compound O=[Si]=O VYPSYNLAJGMNEJ-UHFFFAOYSA-N 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 229910052710 silicon Inorganic materials 0.000 description 1
- 239000010703 silicon Substances 0.000 description 1
- 239000000377 silicon dioxide Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
- G06T2207/10021—Stereoscopic video; Stereoscopic image sequence
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Testing, Inspecting, Measuring Of Stereoscopic Televisions And Televisions (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a binocular 3D distortion rectification method based on a GPU. The method comprises steps of segmenting left and right eye frames by using the GPU, acquiring a main foreground object, and placing objects near a central point on a zero plane; screening and registering the characteristic points of the foreground object; rotating and scaling the left and right eye frames; regulating a parallax error according to translation quantity. Compared with original left and right eye original frames, the processed left and right eye frames are decreased in image edge distortion. A problem is solved that the left and right eye frames are not in the same plane previously. The letter box of the image edge does not exist fundamentally. The parallax error of the left and right eye frames is reduced. The target frames are significantly improved in watching comfortableness than the original frames.
Description
Technical field
The present invention relates to a kind of binocular 3D distortion correction method based on GPU processor.
Background technology
In recent years, the micro display technology with liquid crystal on silicon, silica-based OLED as Typical Representative is developed rapidly, meanwhile,
3D imaging and Display Technique are developed rapidly the most in the near future, and corresponding product is at aspects such as military affairs, field detection, medical science
The most universal, particularly benefit from large screen flat plate display (display panels, PDP display floater) and popularize rapidly,
3D imaging device in civil area is the most of common occurrence.Research according to the 3D image disparity principle to human eye draws, binocular 3D
The most real 3D imaging means of imaging technique.
Summary of the invention
It is an object of the invention to overcome the deficiencies in the prior art, it is provided that a kind of binocular 3D distortion correction side based on GPU processor
Method, uses GPU to split right and left eyes picture, obtains main foreground object, central point adnexa object is placed in zero plane；
Foreground object is carried out characteristic point screening registration；Right and left eyes picture is rotated and scales correction；Parallax is regulated according to translational movement.
It is an object of the invention to be achieved through the following technical solutions: a kind of binocular 3D distortion correction side based on GPU processor
Method, it comprises the steps:
S1: right and left eyes picture LImg1 and RImg1 of video camera shooting is carried out camera calibration, calculates the inside of video camera
Parameter, obtains radial distortion and tangential distortion information, and computing formula is as follows:
Wherein,(xd,yd) it is preferred view coordinate, (xu,yu) it is actual projection coordinate, (k1,k2) for imaging
Machine radial distortion, (k3,k4) it is tangential distortion；
S2: be up in GPU by right and left eyes picture LImg1 and RImg1, uses the formula that S1 obtains, pixel-by-pixel to original left
Right eye picture carries out the distortion correction of inner parameter, obtains LImg2 and RImg2；
S3: right and left eyes picture LImg2 and RImg2 after distortion correction is carried out in GPU edge and carries contours extract and image divides
Cut, obtain main foreground object sequence LObj and the RObj of right and left eyes picture respectively；
S4: main foreground object sequence LObj and RObj to right and left eyes picture carry out feature extraction in GPU, respectively obtain
The rotation of LObj and RObj, Pan and Zoom information；
S5: calculate LObj and RObj average translation amount, and then calculate shooting 3D video pictures video camera between relative away from
From；
S6: average according to LObj and RObj rotates and scalability information, carries out right and left eyes picture LImg2 and RImg2 in opposite directions
Rotation and scaling correction, thus by right and left eyes picture rotate to same angle and ensure the main foreground object of right and left eyes picture big
Little unanimously, obtain right and left eyes correction coupling after picture LImg3 and RImg3；
S7: according to correction result screen LImg3 and RImg3, calculate the main front scenery that distance central point in right and left eyes is nearest
Body is as video camera zero plane, and foundation right and left eyes optimal parallax empirical value between 2% to 3%, during in conjunction with 3D video capture
The distance of right and left eyes video camera, with reference to the average translation amount of LObj and RObj, is adjusted the parallax of right and left eyes picture,
After obtain right and left eyes picture LImgD and RImgD that be disposed.
Camera calibration described in step S1 uses the image containing chessboard or grid lines carry out automatic Calibration or use manually side
Formula is selected a plurality of parallel lines and is demarcated.
In step S3, image segmentation uses the image segmentation algorithm that the applicable GPU of limited Clustering calculates, and this algorithm includes as follows
Process step:
A. pending image is up in GPU be denoted as ImgA, to each pixel in ImgA from the upper left corner to the lower right corner according to
Ranks order starts assignment successively from 0, will be denoted as InfoA in this information upstream to GPU；
B. ImgA image is divided into the subimage block of 8*8, utilize the concurrency of GPU simultaneously to subimage block according to color and
Range information carries out sub-cluster, and by cluster result record in InfoA；
The subimage block of the ImgA c. previous step obtained up and down four be one group, regard new subimage block as, according to
InfoA is merged by subimage block boundary pixel value information；
The new subimage block of the ImgA obtained after being d. combined carries out sub-cluster according to color and range information, and by cluster result
Record is in InfoA；
E. it return back to step c order perform, until whole image ImgA only one of which subimage block, then stop cluster, perform
Step f；
F. information search 8 neighborhood to InfoA record, carries out merger, then final InfoA note by isolated point or fritter
Record is exactly the image segmentation result of ImgA.
The SURF algorithm using GPU to optimize in step S4, extracts the characteristic point of Lobj and Robj, uses Euclidean distance to calculate
Obtain the similarity of characteristic point, be ranked up according to similarity, take front 1/2nd characteristic points and registrate, calculated by RANSAC
The characteristic point of pairing is tested screening by method, calculates the rotation of Lobj and Robj, Pan and Zoom information.
The object that during described calculating scalability information, selected distance picture central point is nearest calculates as standard, accelerates scaling letter
The calculating of breath.
The invention has the beneficial effects as follows: the invention provides a kind of binocular 3D distortion correction method based on GPU processor, use
Right and left eyes picture is split by GPU, obtains main foreground object, and central point adnexa object is placed in zero plane；To front scenery
Body carries out characteristic point screening registration；Right and left eyes picture is rotated and scales correction；Parallax is regulated according to translational movement.
Accompanying drawing explanation
Fig. 1 is the flow chart of the inventive method；
Fig. 2 is binocular 3D camera imaging model schematic diagram.
Detailed description of the invention
Technical scheme is described in further detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to following institute
State.
Being illustrated in figure 2 binocular 3D camera imaging model schematic diagram, in Fig. 1, C1 and C2 is the angle point of right and left eyes video camera
Position, (x1, y1, z1) and (x2, y2, z2) be respectively right and left eyes video camera camera coordinates system, (x1, OI 1, y1) and
(x2, OI2, y2) is respectively the image plane of right and left eyes video camera, and the point (Xw, Yw, Zw) in world coordinate system is at right and left eyes video camera
The projection of image plane is respectively (X1, Y1) and (X2, Y2).
Technical scheme, based on Fig. 2 institute representation model, is illustrated.Technical scheme realizes, by following
Several steps complete.
As it is shown in figure 1, a kind of binocular 3D distortion correction method based on GPU processor, it comprises the steps:
S1: input color format be left eye raw frames LImg1 of BGRA and right eye raw frames RImg1 to internal memory, to shooting
Right and left eyes picture LImg1 and RImg1 of machine shooting carries out camera calibration, calculates the inner parameter of video camera, obtains radially
Distortion and tangential distortion information, camera calibration uses the image containing chessboard or grid lines carry out automatic Calibration or use manually side
Formula is selected a plurality of parallel lines and is demarcated, and computing formula is as follows:
Wherein,(xd,yd) it is preferred view coordinate, (xu,yu) it is actual projection coordinate, (k1,k2) for imaging
Machine radial distortion, (k3,k4) it is tangential distortion；
S2: be up in GPU by right and left eyes picture LImg1 and RImg1, uses the formula that S1 obtains, pixel-by-pixel to original left
Right eye picture LImg1 and RImg1 carries out the distortion correction of inner parameter, the right and left eyes picture LImg2 after being corrected and
RImg2；
S3: use the GPU image segmentation algorithm of limited clustering method to right and left eyes picture LImg2 and RImg2 after distortion correction
GPU is carried out edge carry contours extract and image segmentation, respectively obtain right and left eyes picture main foreground object sequence LObj and
RObj；
S4: right and left eyes foreground object sequence LObj and RObj that step 3 is obtained by use SURF algorithm carry out feature extraction,
To right and left eyes foreground object characteristic of correspondence point, re-use Euclidean distance and be calculated characteristic point similarity, and arrange according to similarity
Sequence, before taking similarity afterwards, the characteristic point of 1/2nd registrates.It is then used by RANSAC algorithm the characteristic point of pairing is carried out
Inspection screening, finally calculates right and left eyes foreground object sequence LObj and the rotation of RObj, Pan and Zoom information；Calculate contracting
The object that when putting information, selected distance picture central point is nearest calculates as standard, accelerates the calculating of scalability information；
S5: according to right and left eyes foreground object sequence LObj and the rotation of RObj, Pan and Zoom information, is calculated average rotation
Turn, average translation and average scalability information, calculate the binocular camera of shooting right and left eyes raw frames according to average translation information
Between relative distance；
S6: reversely rotate according to right and left eyes picture LImg2 and RImg2 after distortion correction and scale, thus by left and right
Eye picture rotates to identical angle position and ensure that the main foreground object of right and left eyes picture is in the same size, obtains right and left eyes school
Picture LImg3 and RImg3 after just mating；
S7: according to correction result screen LImg3 and RImg3, calculate the main front scenery that distance central point in right and left eyes is nearest
Body as video camera zero plane, then according to right and left eyes disparity range 2% to 3% empirical value, integrating step 5 calculates
Distance between the binocular camera of shooting right and left eyes video, with reference to right and left eyes foreground object sequence LObj and the average translation of RObj
Amount, is adjusted right and left eyes picture parallax, finally obtains right and left eyes picture LImgD and RImgD being disposed.
By final right and left eyes picture LImgD and RImgD from GPU up-downgoing to right and left eyes internal storage data, and internal storage data is made
For the purpose of output return.
In step S3, image segmentation uses the image segmentation algorithm that the applicable GPU of limited Clustering calculates.Describe below this
Pending image is up in GPU be denoted as ImgA by the process step of algorithm: a., to each pixel in ImgA from the upper left corner
Start successively assignment according to ranks order from 0 to the lower right corner, this information upstream to GPU will be denoted as InfoA；B. by ImgA
Image is divided into the subimage block of 8*8, utilizes the concurrency of GPU according to color and range information, subimage block to be carried out son simultaneously
Cluster, and by cluster result record in InfoA；The subimage block of the ImgA c. previous step obtained four up and down
It is one group, regards new subimage block as, according to subimage block boundary pixel value information (color and distance), InfoA is closed
And；The new subimage block of the ImgA obtained after being d. combined carries out sub-cluster according to color and range information, and by cluster result
Record is in InfoA；E. it return back to step c order perform, until whole image ImgA only one of which subimage block, then stop
Only cluster, performs step f；F. information search 8 neighborhood to InfoA record, carries out merger by isolated point or fritter,
Then what final InfoA recorded is exactly the image segmentation result of ImgA.
Right and left eyes picture image marginal distortion after the binocular 3D distortion correction method of the present invention is corrected reduces a lot, the most original
Right and left eyes image not have also been obtained solution in conplane problem, then image border problem of black borders is also substantially not present,
Finally contrasting right and left eyes picture parallax again, image parallactic also reduces, and purpose picture is compared raw frames and has been obviously improved viewing
Comfort level.
Claims (5)
1. a binocular 3D distortion correction method based on GPU processor, it is characterised in that: it comprises the steps:
S1: right and left eyes picture LImg1 and RImg1 of video camera shooting is carried out camera calibration, calculates the inside of video camera
Parameter, obtains radial distortion and tangential distortion information, and computing formula is as follows:
Wherein, For preferred view coordinate, (xu,yu) it is actual projection coordinate, (k1,k2) for imaging
Machine radial distortion, (k3,k4) it is tangential distortion；
S2: be up in GPU by right and left eyes picture LImg1 and RImg1, uses the formula that S1 obtains, pixel-by-pixel to original left
Right eye picture carries out the distortion correction of inner parameter, obtains LImg2 and RImg2；
S3: right and left eyes picture LImg2 and RImg2 after distortion correction is carried out in GPU edge and carries contours extract and image divides
Cut, obtain main foreground object sequence LObj and the RObj of right and left eyes picture respectively；
S4: main foreground object sequence LObj and RObj to right and left eyes picture carry out feature extraction in GPU, respectively obtain
The rotation of LObj and RObj, Pan and Zoom information；
S5: calculate LObj and RObj average translation amount, and then calculate shooting 3D video pictures video camera between relative away from
From；
S6: average according to LObj and RObj rotates and scalability information, carries out right and left eyes picture LImg2 and RImg2 in opposite directions
Rotation and scaling correction, thus by right and left eyes picture rotate to same angle and ensure the main foreground object of right and left eyes picture big
Little unanimously, obtain right and left eyes correction coupling after picture LImg3 and RImg3；
S7: according to correction result screen LImg3 and RImg3, calculate the main front scenery that distance central point in right and left eyes is nearest
Body is as video camera zero plane, and foundation right and left eyes optimal parallax empirical value between 2% to 3%, during in conjunction with 3D video capture
The distance of right and left eyes video camera, with reference to the average translation amount of LObj and RObj, is adjusted the parallax of right and left eyes picture,
After obtain right and left eyes picture LImgD and RImgD that be disposed.
A kind of binocular 3D distortion correction method based on GPU processor the most according to claim 1, it is characterised in that:
Camera calibration described in step S1 uses the image containing chessboard or grid lines carry out automatic Calibration or use manual mode choosing
Go out a plurality of parallel lines to demarcate.
A kind of binocular 3D distortion correction method based on GPU processor the most according to claim 1, it is characterised in that:
In step S3, image segmentation uses the image segmentation algorithm that the applicable GPU of limited Clustering calculates, and this algorithm includes following flow process
Step:
A. pending image is up in GPU be denoted as ImgA, to each pixel in ImgA from the upper left corner to the lower right corner according to
Ranks order starts assignment successively from 0, will be denoted as InfoA in this information upstream to GPU；
B. ImgA image is divided into the subimage block of 8*8, utilize the concurrency of GPU simultaneously to subimage block according to color and
Range information carries out sub-cluster, and by cluster result record in InfoA；
The subimage block of the ImgA c. previous step obtained up and down four be one group, regard new subimage block as, according to
InfoA is merged by subimage block boundary pixel value information；
The new subimage block of the ImgA obtained after being d. combined carries out sub-cluster according to color and range information, and by cluster result
Record is in InfoA；
E. it return back to step c order perform, until whole image ImgA only one of which subimage block, then stop cluster, perform
Step f；
F. information search 8 neighborhood to InfoA record, carries out merger, then final InfoA note by isolated point or fritter
Record is exactly the image segmentation result of ImgA.
A kind of binocular 3D distortion correction method based on GPU processor the most according to claim 1, it is characterised in that:
The SURF algorithm using GPU to optimize in step S4, extracts the characteristic point of Lobj and Robj, uses Euclidean distance to be calculated
The similarity of characteristic point, is ranked up according to similarity, takes front 1/2nd characteristic points and registrate, by RANSAC algorithm pair
The characteristic point of pairing is tested screening, calculates the rotation of Lobj and Robj, Pan and Zoom information.
A kind of binocular 3D distortion correction method based on GPU processor the most according to claim 4, it is characterised in that:
The object that during described calculating scalability information, selected distance picture central point is nearest calculates as standard, accelerates scalability information
Calculate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610318480.5A CN106023170B (en) | 2016-05-13 | 2016-05-13 | A kind of binocular 3D distortion correction methods based on GPU processors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610318480.5A CN106023170B (en) | 2016-05-13 | 2016-05-13 | A kind of binocular 3D distortion correction methods based on GPU processors |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106023170A true CN106023170A (en) | 2016-10-12 |
| CN106023170B CN106023170B (en) | 2018-10-23 |
Family
ID=57100077
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610318480.5A Active CN106023170B (en) | 2016-05-13 | 2016-05-13 | A kind of binocular 3D distortion correction methods based on GPU processors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106023170B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107154027A (en) * | 2017-04-17 | 2017-09-12 | 深圳大学 | Compensation method and device that a kind of fault image restores |
| CN107481267A (en) * | 2017-08-14 | 2017-12-15 | 华南理工大学 | A kind of shooting projection interactive system and method based on binocular vision |
| CN107492080A (en) * | 2017-09-25 | 2017-12-19 | 天津智慧视通科技有限公司 | Exempt from calibration easily monocular lens image radial distortion antidote |
| CN109842756A (en) * | 2017-11-28 | 2019-06-04 | 东莞市普灵思智能电子有限公司 | A kind of method and system of lens distortion correction and feature extraction |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104010178A (en) * | 2014-06-06 | 2014-08-27 | 深圳市墨克瑞光电子研究院 | Binocular image parallax adjusting method and device and binocular camera |
| US20160105662A1 (en) * | 2014-10-08 | 2016-04-14 | Samsung Display Co., Ltd. | Three-dimensional glasses and method of driving the same |
-
2016
- 2016-05-13 CN CN201610318480.5A patent/CN106023170B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104010178A (en) * | 2014-06-06 | 2014-08-27 | 深圳市墨克瑞光电子研究院 | Binocular image parallax adjusting method and device and binocular camera |
| US20160105662A1 (en) * | 2014-10-08 | 2016-04-14 | Samsung Display Co., Ltd. | Three-dimensional glasses and method of driving the same |
Non-Patent Citations (4)
| Title |
|---|
| XIE RUOBING 等: "The Research on Binocular Stereo Video Imaging and Display System Based on Low-light CMOS", 《APPLIED OPTICS AND PHOTONICS CHINA》 * |
| 刘春群 等: "基于特征点图像配准方法的应用研究", 《科技与企业》 * |
| 咋了啦: "相机标定原理-标定原理介绍", 《CSDN》 * |
| 张开玉: "基于改进SURF的双目立体视觉三维重建", 《中国优秀硕士学位论文全文数据库 信息科技辑（月刊）》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107154027A (en) * | 2017-04-17 | 2017-09-12 | 深圳大学 | Compensation method and device that a kind of fault image restores |
| CN107154027B (en) * | 2017-04-17 | 2020-07-07 | 深圳大学 | Compensation method and device for restoration of distorted image |
| CN107481267A (en) * | 2017-08-14 | 2017-12-15 | 华南理工大学 | A kind of shooting projection interactive system and method based on binocular vision |
| CN107492080A (en) * | 2017-09-25 | 2017-12-19 | 天津智慧视通科技有限公司 | Exempt from calibration easily monocular lens image radial distortion antidote |
| CN107492080B (en) * | 2017-09-25 | 2021-03-23 | 杭州云栖智慧视通科技有限公司 | Calibration-free convenient monocular head image radial distortion correction method |
| CN109842756A (en) * | 2017-11-28 | 2019-06-04 | 东莞市普灵思智能电子有限公司 | A kind of method and system of lens distortion correction and feature extraction |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106023170B (en) | 2018-10-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3163535B1 (en) | Wide-area image acquisition method and device | |
| CN109314753B (en) | Method and computer-readable storage medium for generating intermediate views using optical flow | |
| CN107358577B (en) | Rapid splicing method of cubic panoramic image | |
| CN110211043A (en) | A kind of method for registering based on grid optimization for Panorama Mosaic | |
| CN111047510A (en) | Large-field-angle image real-time splicing method based on calibration | |
| CN103337094A (en) | Method for realizing three-dimensional reconstruction of movement by using binocular camera | |
| CN110956661B (en) | Method for calculating dynamic pose of visible light and infrared camera based on bidirectional homography matrix | |
| CN110910431B (en) | Multi-view three-dimensional point set recovery method based on monocular camera | |
| CN102982524B (en) | Splicing method for corn ear order images | |
| CN106023170A (en) | Binocular 3D distortion rectification method based on GPU | |
| CN105631859B (en) | Three-degree-of-freedom bionic stereo visual system | |
| CN110099268B (en) | Blind area perspective display method with natural color matching and natural display area fusion | |
| CN103313081A (en) | Image processing apparatus and method | |
| CN108181319A (en) | A kind of laying dust detecting device and method based on stereoscopic vision | |
| CN107016707B (en) | A kind of integration imaging super large three-dimensional scenic shooting method for correcting image | |
| CN109493282A (en) | A kind of stereo-picture joining method for eliminating movement ghost image | |
| WO2020151268A1 (en) | Generation method for 3d asteroid dynamic map and portable terminal | |
| CN105606123A (en) | Method for automatic correction of digital ground elevation model for low-altitude aerial photogrammetry | |
| CN107330856B (en) | Panoramic imaging method based on projective transformation and thin plate spline | |
| Gao et al. | Design of signal processing pipeline for stereoscopic cameras | |
| CN113096016A (en) | Low-altitude aerial image splicing method and system | |
| CN110430400A (en) | A kind of ground level method for detecting area of the movable video camera of binocular | |
| EP3229106A1 (en) | Efficient determination of optical flow between images | |
| CN108830804A (en) | Virtual reality fusion Fuzzy Consistent processing method based on line spread function standard deviation | |
| CN111629194B (en) | Method and system for converting panoramic video into 6DOF video based on neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |