CN108399632A - A kind of RGB-D camera depth image repair methods of joint coloured image - Google Patents

A kind of RGB-D camera depth image repair methods of joint coloured image Download PDF

Info

Publication number
CN108399632A
CN108399632A CN201810174885.5A CN201810174885A CN108399632A CN 108399632 A CN108399632 A CN 108399632A CN 201810174885 A CN201810174885 A CN 201810174885A CN 108399632 A CN108399632 A CN 108399632A
Authority
CN
China
Prior art keywords
depth
rgb
image
camera
cavity
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
Application number
CN201810174885.5A
Other languages
Chinese (zh)
Other versions
CN108399632B (en
Inventor
张毅
戴健春
张瀚
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University of Post and Telecommunications
Original Assignee
Chongqing University of Post and Telecommunications
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Chongqing University of Post and Telecommunications filed Critical Chongqing University of Post and Telecommunications
Priority to CN201810174885.5A priority Critical patent/CN108399632B/en
Publication of CN108399632A publication Critical patent/CN108399632A/en
Application granted granted Critical
Publication of CN108399632B publication Critical patent/CN108399632B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/20—Image enhancement or restoration by the use of local operators
    • G06T5/30—Erosion or dilatation, e.g. thinning
    • G06T5/70—
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/12—Edge-based segmentation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
    • 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
    • G06T2207/10024—Color image

Abstract

The present invention relates to a kind of RGB D camera depth image repair methods of joint coloured image, belong to depth image and repair field.This method includes：Colored, depth camera is demarcated using Zhang Shi standardizations, obtains the inside and outside parameter of camera；Colored, depth camera coordinate alignment is realized according to pinhole imaging system principle and coordinate transform；Obtain that scene is colored, depth image respectively, using depth threshold method by its binaryzation；Judge that the size of empty connected domain has cavity to determine；It carries out expansive working and obtains empty neighborhood；Cavity is divided into according to its size and blocks, is empty in plane by the depth value variance for calculating empty field pixel；Colour consistency is respectively adopted and similitude pair two class cavity in field is repaired；Image after reparation is filtered using part filter method, removes noise.The present invention maintains the raw information of depth image, and improve the remediation efficiency of depth image as far as possible while ensureing that depth image is repaired.

Description

A kind of RGB-D camera depth image repair methods of joint coloured image
Technical field
The invention belongs to depth images to repair field, be related to a kind of RGB-D depth images reparation side of joint coloured image Method.
Background technology
With the development of vision technique, the camera of more and more types is applied in display scene, such as：It is common color Form and aspect machine is used for recognition of face, and CCD camera is used for detection of obstacles, and Kinect is for skeleton tracking etc..Ordinary two dimensional is flat Face camera can obtain scene two-dimensional signal, and for three-dimensional display space, two dimensional surface camera obviously cannot be satisfied reality Application demand, although two or more two dimensional surface camera combinations can be utilized to be obtained using scene three-dimensional information can be achieved, Its is inconvenient to use and computationally intensive.This case is tentatively improved with the appearance of RGB-D cameras, such as Microsoft's publication Kinect very easily obtains depth information while obtaining scene coloured image.With Microsoft Kinect's (natal) Development, depth camera attract the sight of more and more people, depth camera that can be used in human body tracking, and three-dimensional reconstruction is man-machine Interaction, the fields such as SLAM.But since the depth image that RGB-D cameras obtain typically contains more cavity, depth can not be believed Breath is directly applied in practical application, this greatly affected the application range of RGB-D cameras.
Therefore, the deep image information presence of RGB-D acquisitions is solved there is an urgent need for a kind of effective depth image restorative procedure Cavity and the problem of can not directly use.
Invention content
In view of this, the purpose of the present invention is to provide a kind of RGB-D camera depth image repairs of joint coloured image There is cavity for the RGB-D deep image informations obtained and can not directly use, propose a kind of effective depth in method Spend image repair method.
In order to achieve the above objectives, the present invention provides the following technical solutions：
A kind of RGB-D camera depth image repair methods of joint coloured image, include the following steps：
S1：Colour imagery shot and depth camera are demarcated using classical Zhang Shi standardizations, obtain the interior of camera Outer parameter；
S2：Colour imagery shot and the alignment of depth camera coordinate are realized according to pinhole imaging system principle and coordinate transform；
S3：Scene coloured image and depth image are obtained respectively, and depth image is carried out by binaryzation according to depth threshold method；
S4：Judge that the size of depth image cavity connected domain has cavity to determine；
S5：Expansive working is carried out to depth image and obtains empty neighborhood；
S6：The depth value variance for calculating empty field pixel, cavity is divided into according to the size of variance block cavity and It is empty in plane；
S7：Colour consistency is respectively adopted according to different empty types and similitude pair two class cavity in field is repaiied It is multiple；
S8：Image after reparation is filtered using the method for part filter, while ensureing depth value primitiveness Remove noise.
Further, the step S2 specifically includes following steps：
S21：The spin matrix R of color camera is obtained through step S1rgbWith translation vector Trgb, the spin matrix of depth camera RirWith translation vector Tir；If P is a bit under world coordinate system, PrgbWith PirRespectively P is under color camera and depth camera Projection coordinate's point, relational expression is obtained by national forest park in Xiaokeng：
Prgb=RrgbP+Trgb
Pir=RirP+Tir
S22：If prgbAnd pirRespectively projection coordinate of this in RGB image planes and depth image plane, by camera Join matrix ArgbAnd AirIt obtains：
prgb=ArgbPrgb
pir=AirPir
S23：The coordinate of depth camera is different with the coordinate of RGB cameras, PrgbAnd PirWith a spin matrix R and one A translation matrix T is connected, and the relationship of the two is：
Prgb=RPir+T
S24：P is eliminated by S21 equatioies to obtain：
S25：Simultaneous S23 and S24 equation respective items are equal to obtain R and T：
S26：Simultaneous S22 and S23 equation, finally obtains the correspondence of color image pixel point and depth image pixel For：
Further, the step S3 specifically includes following steps：
S31：According to after alignment color camera and depth camera simultaneously obtain scene image；
S32：Original depth data is quantified, depth value is transformed into gray level 0~255, quantitative formula is as follows：
Wherein depth is the depth value after quantization, DpFor the depth value of each pixel, DmaxFor maximum depth value；
S33：Binaryzation is carried out according to depth threshold method to the depth image of acquisition, formula is expressed as follows：
Wherein I (x, y) is the image after binaryzation, and depth (x, y) represents the depth value after pixel quantization, DthrIt is two Value depth threshold.
Further, the step S4 is specifically included：After S3 binary conversion treatments, the size of empty connected domain is judged, with Cavity and noise spot when differentiation；If connected domain judgement is cavity, continue subsequent processing；If it is not, then without subsequent Processing.
Further, the step S5 specifically includes following steps：
S51：Expansive working H is carried out to the cavity that S4 judgesep；
S52：Obtain cavity field Hnp, expression formula is as follows：
Hnp=Hep-Ho
Wherein HoFor hole region.
Further, the step S6 specifically includes following steps：
S61：According to the empty neighborhood that S5 is obtained, the variance Δ of depth value after pixel quantization in neighborhood is calculateddepth, calculate Formula is as follows：
Wherein DiIndicate the depth value of i pixels in empty field,Indicate that empty field depth value mean value, i indicate pixel Point number, N indicate empty neighborhood territory pixel point total number；
S62：The neighborhood variance Δ being calculated using S61depthWith empty judgment threshold ΔthIt compares, if empty neighborhood Depth variance ΔdepthLess than ΔthThen it is judged as cavity in plane, if it is greater than or equal to ΔthThen it is judged as blocking cavity.
Further, the step S7 is specifically included：It is empty for being judged as in plane, i.e., the cavity be object plane it is interior by In the cavity do not reflected infrared light and generated, due to being present in a plane, the depth value of the hole region and its neighborhood The depth of pixel is similar, is repaired using the alignment of field pixel depth value；And for being judged as blocking cavity, be due to Object mutually blocks and the cavity that generates, and this kind of cavity is repaired and is then repaired using coloured image, that is, finds out empty neighbour The depth value for the pixel being closer to color image color in hole region in domain is repaired.
Further, the step S8 is specifically included：Since there are still tiny noises for the cavity after S7 is repaired, in order to the greatest extent It can guarantee the primitiveness of depth data, while noise can be removed, image filtering is carried out using local top filtering method.
The beneficial effects of the present invention are：Cavity is divided into two classes by the present invention according to empty neighborhood variance threshold values method, joint Field similitude is respectively adopted in coloured image and colour consistency carries out empty reparation, and the image after cavity is repaired is filtered using part The method of wave removes noise.Compared to traditional rough restorative procedure using bilateral filtering algorithm, the present invention is ensureing depth map As the validity repaired ensure that the primitiveness of deep image information as far as possible simultaneously.
Description of the drawings
In order to keep the purpose of the present invention, technical solution and advantageous effect clearer, the present invention provides following attached drawing and carries out Explanation：
Fig. 1 is the system flow chart of the present invention；
Fig. 2 is empty neighborhood schematic diagram calculation；
Fig. 3 is the schematic diagram for blocking cavity formation.
Specific implementation mode
Below in conjunction with attached drawing, the preferred embodiment of the present invention is described in detail.
Fig. 1 is the system flow chart of the present invention, as shown in Figure 1, a kind of RGB-D camera depth images of joint coloured image Restorative procedure first has to realize being aligned for color camera and depth camera coordinate, according to depth threshold method to the depth after quantization Image carries out binaryzation, and empty neighborhood is obtained by expansive working, calculates neighborhood variance to judge empty type, for different Field similitude is respectively adopted in cavity and colour consistency is repaired, and finally carries out image filter using local bilateral filtering method Wave.
This method specifically includes following steps：
S1：Colour imagery shot and depth camera are demarcated using classical Zhang Shi standardizations, obtain the interior of camera Outer parameter；
S2：Colour imagery shot and the alignment of depth camera coordinate are realized according to pinhole imaging system principle and coordinate transform；
S3：Scene coloured image and depth image are obtained respectively, and depth image is carried out by binaryzation according to depth threshold method；
S4：Judge that the size of depth image cavity connected domain has cavity to determine：After S3 binary conversion treatments, judge The size of empty connected domain, cavity and noise spot when distinguishing；If connected domain judgement is cavity, continue subsequent processing；If It is not, then without subsequent processing.
S5：Expansive working is carried out to depth image and obtains empty neighborhood；
S6：The depth value variance for calculating empty field pixel, cavity is divided into according to the size of variance block cavity and It is empty in plane；
S7：Colour consistency is respectively adopted according to different empty types and similitude pair two class cavity in field is repaiied It is multiple；
Empty for being judged as in plane, i.e., the cavity is the sky that is generated in object plane due to not reflecting infrared light Hole, due to being present in a plane, the depth value of the hole region is similar to the depth of its neighborhood territory pixel point, utilizes field The alignment of pixel depth value is repaired；And for being judged as blocking cavity, be due to the cavity that object mutually blocks and generates, It is then repaired, that is, found out in empty neighborhood and in hole region using coloured image as shown in figure 3, being repaired for this kind of cavity The depth value for the pixel that color image color is closer to is repaired.
S8：Image after reparation is filtered using the method for part filter, while ensureing depth value primitiveness Remove noise；
Due to the cavity after S7 is repaired, there are still tiny noises, in order to ensure the primitiveness of depth data as far as possible, Noise can be removed simultaneously, image filtering is carried out using local top filtering method.
Step S2 specifically includes following steps：
S21：The spin matrix R of color camera is obtained through step S1rgbWith translation vector Trgb, the spin matrix of depth camera RirWith translation vector Tir；If P is a bit under world coordinate system, PrgbWith PirRespectively P is under color camera and depth camera Projection coordinate's point, relational expression is obtained by national forest park in Xiaokeng：
Prgb=RrgbP+Trgb
Pir=RirP+Tir
S22：If pgbrAnd pirRespectively projection coordinate of this in RGB image planes and depth image plane, by camera Join matrix ArgbAnd AirIt obtains：
prgb=ArgbPrgb
pir=AirPir
S23：The coordinate of depth camera is different with the coordinate of RGB cameras, PrgbAnd PirWith a spin matrix R and one A translation matrix T is connected, and the relationship of the two is：
Prgb=RPir+T
S24：P is eliminated by S21 equatioies to obtain：
S25：Simultaneous S23 and S24 equation respective items are equal to obtain R and T：
S26：Simultaneous S22 and S23 equation, finally obtains the correspondence of color image pixel point and depth image pixel For：
Step S3 specifically includes following steps：
S31：According to after alignment color camera and depth camera simultaneously obtain scene image；
S32：Original depth data is quantified, depth value is transformed into gray level 0~255, quantitative formula is as follows：
Wherein depth is the depth value after quantization, DpFor the depth value of each pixel, DmaxFor maximum depth value；
S33：Binaryzation is carried out according to depth threshold method to the depth image of acquisition, formula is expressed as follows：
Wherein I (x, y) is the image after binaryzation, and depth (x, y) represents the depth value after pixel quantization, DthrIt is two Value depth threshold.
Step S5 specifically includes following steps：
S51：Expansive working H is carried out to the cavity that S4 judgesep；
S52：Obtain cavity field Hnp, expression formula is as follows：
Hnp=Hep-Ho
Wherein HoFor hole region.As shown in Fig. 2, " ten " the character form structure element using 3 × 3 carries out expansion behaviour to cavity Make, the image after expansion, which is subtracted original empty image, can be obtained empty neighborhood.
Step S6 specifically includes following steps：
S61：According to the empty neighborhood that S5 is obtained, the variance Δ of depth value after pixel quantization in neighborhood is calculateddepth, calculate Formula is as follows：
Wherein DiIndicate the depth value of i pixels in empty field,Indicate that empty field depth value mean value, i indicate pixel Point number, N indicate empty neighborhood territory pixel point total number；
S62：The neighborhood variance Δ being calculated using S61depthWith empty judgment threshold ΔthIt compares, if empty neighborhood Depth variance ΔdepthLess than ΔthThen it is judged as cavity in plane, if it is greater than or equal to ΔthThen it is judged as blocking cavity.
Finally illustrate, preferred embodiment above is merely illustrative of the technical solution of the present invention and unrestricted, although logical It crosses above preferred embodiment the present invention is described in detail, however, those skilled in the art should understand that, can be Various changes are made to it in form and in details, without departing from claims of the present invention limited range.

Claims (8)

1. a kind of RGB-D camera depth image repair methods of joint coloured image, which is characterized in that this method includes following step Suddenly：
S1：Colour imagery shot and depth camera are demarcated using classical Zhang Shi standardizations, obtain the inside and outside ginseng of camera Number；
S2：Colour imagery shot and the alignment of depth camera coordinate are realized according to pinhole imaging system principle and coordinate transform；
S3：Scene coloured image and depth image are obtained respectively, and depth image is carried out by binaryzation according to depth threshold method；
S4：Judge that the size of depth image cavity connected domain has cavity to determine；
S5：Expansive working is carried out to depth image and obtains empty neighborhood；
S6：Cavity is divided into according to the size of variance and blocks cavity and plane by the depth value variance for calculating empty field pixel Interior cavity；
S7：Colour consistency is respectively adopted according to different empty types and similitude pair two class cavity in field is repaired；
S8：Image after reparation is filtered using the method for part filter, is removed while ensureing depth value primitiveness Noise.
2. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S2 specifically includes following steps：
S21：The spin matrix R of color camera is obtained through step S1rgbWith translation vector Trgb, the spin matrix R of depth camerairWith Translation vector Tir；If P is a bit under world coordinate system, PrgbWith PirThrowings of the respectively P under color camera and depth camera Shadow coordinate points obtain relational expression by national forest park in Xiaokeng：
Prgb=RrgbP+Trgb
Pir=RirP+Tir
S22：If prgbAnd pirRespectively projection coordinate of this in RGB image planes and depth image plane, by the internal reference square of camera Battle array ArgbAnd AirIt obtains：
prgb=ArgbPrgb
pir=AirPir
S23：The coordinate of depth camera is different with the coordinate of RGB cameras, PrgbAnd PirIt is flat with a spin matrix R and one It moves matrix T to connect, the relationship of the two is：
Prgb=RPir+T
S24：P is eliminated by S21 equatioies to obtain：
S25：Simultaneous S23 and S24 equation respective items are equal to obtain R and T：
S26：Simultaneous S22 and S23 equation, finally obtains color image pixel point and the correspondence of depth image pixel is：
3. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S3 specifically includes following steps：
S31：According to after alignment color camera and depth camera simultaneously obtain scene image；
S32：Original depth data is quantified, depth value is transformed into gray level 0~255, quantitative formula is as follows：
Wherein depth is the depth value after quantization, DpFor the depth value of each pixel, DmaxFor maximum depth value；
S33：Binaryzation is carried out according to depth threshold method to the depth image of acquisition, formula is expressed as follows：
Wherein I (x, y) is the image after binaryzation, and depth (x, y) represents the depth value after pixel quantization, DthrFor binaryzation Depth threshold.
4. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S4 is specifically included：After S3 binary conversion treatments, judge the size of empty connected domain, when distinguishing cavity and Noise spot；If connected domain judgement is cavity, continue subsequent processing；If it is not, then without subsequent processing.
5. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S5 specifically includes following steps：
S51：Expansive working H is carried out to the cavity that S4 judgesep；
S52：Obtain cavity field Hnp, expression formula is as follows：
Hnp=Hep-Ho
Wherein HoFor hole region.
6. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S6 specifically includes following steps：
S61：According to the empty neighborhood that S5 is obtained, the variance Δ of depth value after pixel quantization in neighborhood is calculateddepth, calculation formula It is as follows：
Wherein DiIndicate the depth value of i pixels in empty field,Indicate that empty field depth value mean value, i indicate that pixel is compiled Number, N indicates empty neighborhood territory pixel point total number；
S62：The neighborhood variance Δ being calculated using S61depthWith empty judgment threshold ΔthIt compares, if empty neighborhood depth Variance ΔdepthLess than ΔthThen it is judged as cavity in plane, if it is greater than or equal to ΔthThen it is judged as blocking cavity.
7. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S7 is specifically included：Empty for being judged as in plane, i.e., the cavity is infrared due to not reflecting in object plane Light and the cavity generated, due to being to be present in a plane, the depth of the depth value of the hole region and its neighborhood territory pixel point It is similar, it is repaired using the alignment of field pixel depth value；And for being judged as blocking cavity, it is since object mutually blocks And the cavity generated, this kind of cavity is repaired and is then repaired using coloured image, that is, is found out in empty neighborhood and hole area The depth value for the pixel that color image color is closer in domain is repaired.
8. a kind of RGB-D camera depth image repair methods of joint coloured image according to claim 1, feature exist In the step S8 is specifically included：Due to the cavity after S7 is repaired, there are still tiny noises, in order to ensure depth as far as possible The primitiveness of data, while noise can be removed, image filtering is carried out using local top filtering method.
CN201810174885.5A 2018-03-02 2018-03-02 RGB-D camera depth image restoration method based on color image combination Active CN108399632B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810174885.5A CN108399632B (en) 2018-03-02 2018-03-02 RGB-D camera depth image restoration method based on color image combination

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810174885.5A CN108399632B (en) 2018-03-02 2018-03-02 RGB-D camera depth image restoration method based on color image combination

Publications (2)

Publication Number Publication Date
CN108399632A true CN108399632A (en) 2018-08-14
CN108399632B CN108399632B (en) 2021-06-15

Family

ID=63091760

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810174885.5A Active CN108399632B (en) 2018-03-02 2018-03-02 RGB-D camera depth image restoration method based on color image combination

Country Status (1)

Country Link
CN (1) CN108399632B (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109345580A (en) * 2018-10-23 2019-02-15 北京字节跳动网络技术有限公司 Method and apparatus for handling image
CN109360230A (en) * 2018-11-08 2019-02-19 武汉库柏特科技有限公司 A kind of method for registering images and system based on 2D camera Yu 3D camera
CN109636732A (en) * 2018-10-24 2019-04-16 深圳先进技术研究院 A kind of empty restorative procedure and image processing apparatus of depth image
CN109685732A (en) * 2018-12-18 2019-04-26 重庆邮电大学 A kind of depth image high-precision restorative procedure captured based on boundary
CN109741405A (en) * 2019-01-21 2019-05-10 同济大学 A kind of depth information acquisition system based on dual structure light RGB-D camera
CN110866882A (en) * 2019-11-21 2020-03-06 湖南工程学院 Layered joint bilateral filtering depth map restoration algorithm based on depth confidence
WO2020118896A1 (en) * 2018-12-10 2020-06-18 上海奕瑞光电子科技股份有限公司 Image repair method, image repair system, and flat panel detector
CN111524193A (en) * 2020-04-17 2020-08-11 西安交通大学 Method and device for measuring two-dimensional size of object
CN111626086A (en) * 2019-02-28 2020-09-04 北京市商汤科技开发有限公司 Living body detection method, living body detection device, living body detection system, electronic device, and storage medium
CN111739031A (en) * 2020-06-19 2020-10-02 华南农业大学 Crop canopy segmentation method based on depth information
CN111754713A (en) * 2019-03-28 2020-10-09 杭州海康威视数字技术股份有限公司 Video monitoring method, device and system
CN111862139A (en) * 2019-08-16 2020-10-30 中山大学 Dynamic object parametric modeling method based on color-depth camera
CN112116602A (en) * 2020-08-31 2020-12-22 北京的卢深视科技有限公司 Depth map repairing method and device and readable storage medium
CN112381867A (en) * 2020-11-09 2021-02-19 华南理工大学 Large-area depth image hole automatic filling method for industrial sorting production line
CN112991193A (en) * 2020-11-16 2021-06-18 武汉科技大学 Depth image restoration method, device and computer-readable storage medium
CN113408344A (en) * 2021-05-13 2021-09-17 深圳市捷顺科技实业股份有限公司 Three-dimensional face recognition generation method and related device
CN113628205A (en) * 2021-08-25 2021-11-09 四川大学 Non-contact respiratory frequency detection method based on depth image

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101067557A (en) * 2007-07-03 2007-11-07 北京控制工程研究所 Environment sensing one-eye visual navigating method adapted to self-aid moving vehicle
CN101374242A (en) * 2008-07-29 2009-02-25 宁波大学 Depth map encoding compression method for 3DTV and FTV system
CN101609508A (en) * 2008-06-18 2009-12-23 中国科学院自动化研究所 Sign structure and recognition methods to object identification and orientation information calculating
CN101662694A (en) * 2008-08-29 2010-03-03 深圳华为通信技术有限公司 Method and device for presenting, sending and receiving video and communication system
CN103217111A (en) * 2012-11-28 2013-07-24 西南交通大学 Non-contact contact line geometrical parameter detecting method
CN103561258A (en) * 2013-09-25 2014-02-05 同济大学 Kinect depth video spatio-temporal union restoration method
CN104571482A (en) * 2013-10-22 2015-04-29 中国传媒大学 Digital device control method based on somatosensory recognition
CN107622480A (en) * 2017-09-25 2018-01-23 长春理工大学 A kind of Kinect depth image Enhancement Method

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101067557A (en) * 2007-07-03 2007-11-07 北京控制工程研究所 Environment sensing one-eye visual navigating method adapted to self-aid moving vehicle
CN101609508A (en) * 2008-06-18 2009-12-23 中国科学院自动化研究所 Sign structure and recognition methods to object identification and orientation information calculating
CN101374242A (en) * 2008-07-29 2009-02-25 宁波大学 Depth map encoding compression method for 3DTV and FTV system
CN101662694A (en) * 2008-08-29 2010-03-03 深圳华为通信技术有限公司 Method and device for presenting, sending and receiving video and communication system
CN103217111A (en) * 2012-11-28 2013-07-24 西南交通大学 Non-contact contact line geometrical parameter detecting method
CN103561258A (en) * 2013-09-25 2014-02-05 同济大学 Kinect depth video spatio-temporal union restoration method
CN104571482A (en) * 2013-10-22 2015-04-29 中国传媒大学 Digital device control method based on somatosensory recognition
CN107622480A (en) * 2017-09-25 2018-01-23 长春理工大学 A kind of Kinect depth image Enhancement Method

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
刘田间等: "《一种深度图像修复算法研究》", 《信息技术》 *
吕朝辉等: "《基于Kinect的深度图像修复方法》", 《吉林大学学报》 *
程玉立等: "《面向工业应用的机器人手眼标定与物体定位》", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (27)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109345580A (en) * 2018-10-23 2019-02-15 北京字节跳动网络技术有限公司 Method and apparatus for handling image
CN109345580B (en) * 2018-10-23 2020-03-24 北京字节跳动网络技术有限公司 Method and apparatus for processing image
CN109636732A (en) * 2018-10-24 2019-04-16 深圳先进技术研究院 A kind of empty restorative procedure and image processing apparatus of depth image
CN109360230A (en) * 2018-11-08 2019-02-19 武汉库柏特科技有限公司 A kind of method for registering images and system based on 2D camera Yu 3D camera
WO2020118896A1 (en) * 2018-12-10 2020-06-18 上海奕瑞光电子科技股份有限公司 Image repair method, image repair system, and flat panel detector
CN109685732A (en) * 2018-12-18 2019-04-26 重庆邮电大学 A kind of depth image high-precision restorative procedure captured based on boundary
CN109685732B (en) * 2018-12-18 2023-02-17 重庆邮电大学 High-precision depth image restoration method based on boundary capture
CN109741405A (en) * 2019-01-21 2019-05-10 同济大学 A kind of depth information acquisition system based on dual structure light RGB-D camera
CN109741405B (en) * 2019-01-21 2021-02-02 同济大学 Depth information acquisition system based on dual structured light RGB-D camera
CN111626086A (en) * 2019-02-28 2020-09-04 北京市商汤科技开发有限公司 Living body detection method, living body detection device, living body detection system, electronic device, and storage medium
CN111754713B (en) * 2019-03-28 2021-12-14 杭州海康威视数字技术股份有限公司 Video monitoring method, device and system
CN111754713A (en) * 2019-03-28 2020-10-09 杭州海康威视数字技术股份有限公司 Video monitoring method, device and system
CN111862139A (en) * 2019-08-16 2020-10-30 中山大学 Dynamic object parametric modeling method based on color-depth camera
CN111862139B (en) * 2019-08-16 2023-08-18 中山大学 Dynamic object parametric modeling method based on color-depth camera
CN110866882B (en) * 2019-11-21 2021-09-07 湖南工程学院 Layered joint bilateral filtering depth map repairing method based on depth confidence
CN110866882A (en) * 2019-11-21 2020-03-06 湖南工程学院 Layered joint bilateral filtering depth map restoration algorithm based on depth confidence
CN111524193A (en) * 2020-04-17 2020-08-11 西安交通大学 Method and device for measuring two-dimensional size of object
CN111524193B (en) * 2020-04-17 2022-05-03 西安交通大学 Method and device for measuring two-dimensional size of object
CN111739031A (en) * 2020-06-19 2020-10-02 华南农业大学 Crop canopy segmentation method based on depth information
CN111739031B (en) * 2020-06-19 2023-09-26 华南农业大学 Crop canopy segmentation method based on depth information
CN112116602A (en) * 2020-08-31 2020-12-22 北京的卢深视科技有限公司 Depth map repairing method and device and readable storage medium
CN112381867B (en) * 2020-11-09 2023-09-05 华南理工大学 Automatic filling method for large-area depth image cavity of industrial sorting assembly line
CN112381867A (en) * 2020-11-09 2021-02-19 华南理工大学 Large-area depth image hole automatic filling method for industrial sorting production line
CN112991193A (en) * 2020-11-16 2021-06-18 武汉科技大学 Depth image restoration method, device and computer-readable storage medium
CN113408344A (en) * 2021-05-13 2021-09-17 深圳市捷顺科技实业股份有限公司 Three-dimensional face recognition generation method and related device
CN113628205A (en) * 2021-08-25 2021-11-09 四川大学 Non-contact respiratory frequency detection method based on depth image
CN113628205B (en) * 2021-08-25 2022-05-20 四川大学 Non-contact respiratory frequency detection method based on depth image

Also Published As

Publication number Publication date
CN108399632B (en) 2021-06-15

Similar Documents

Publication Publication Date Title
CN108399632A (en) A kind of RGB-D camera depth image repair methods of joint coloured image
CN103868460B (en) Binocular stereo vision method for automatic measurement based on parallax optimized algorithm
WO2021120406A1 (en) Infrared and visible light fusion method based on saliency map enhancement
CN108898575B (en) Novel adaptive weight stereo matching method
CN105225230B (en) A kind of method and device of identification foreground target object
CN103561258B (en) Kinect depth video spatio-temporal union restoration method
Hernandez et al. Laser scan quality 3-d face modeling using a low-cost depth camera
CN106991663A (en) A kind of under water colour-image reinforcing method theoretical based on dark
CN105447838A (en) Method and system for infrared and low-level-light/visible-light fusion imaging
CN104299220B (en) A kind of method that cavity in Kinect depth image carries out real-time filling
CN108876856A (en) A kind of heavy construction fire fire source recognition positioning method and system
CN109118463B (en) SAR image and optical image fusion method based on HSL and image entropy
CN103034330A (en) Eye interaction method and system for video conference
CN103500440A (en) Method for eliminating cloud and haze of atmospheric degraded image
CN111563852A (en) Dark channel prior defogging method based on low-complexity MF
CN106846258A (en) A kind of single image to the fog method based on weighted least squares filtering
Gao et al. Haze filtering with aerial perspective
CN117036641A (en) Road scene three-dimensional reconstruction and defect detection method based on binocular vision
CN106780384A (en) A kind of the real-time of cold light source abdominal cavity image parameters self adaptation that be applicable goes smog method
CN112802186B (en) Dynamic scene real-time three-dimensional reconstruction method based on binarization characteristic coding matching
CN110223240A (en) Image defogging method, system and storage medium based on color decaying priori
CN105118032B (en) A kind of wide method for dynamically processing of view-based access control model system
CN116757949A (en) Atmosphere-ocean scattering environment degradation image restoration method and system
US20210241430A1 (en) Methods, devices, and computer program products for improved 3d mesh texturing
Xue et al. Optimization of Plane Image Color Enhancement Processing Based on Computer Vision Virtual Reality

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