CN109493282A - A kind of stereo-picture joining method for eliminating movement ghost image - Google Patents
A kind of stereo-picture joining method for eliminating movement ghost image Download PDFInfo
- Publication number
- CN109493282A CN109493282A CN201811389589.3A CN201811389589A CN109493282A CN 109493282 A CN109493282 A CN 109493282A CN 201811389589 A CN201811389589 A CN 201811389589A CN 109493282 A CN109493282 A CN 109493282A
- Authority
- CN
- China
- Prior art keywords
- width
- image
- view
- right view
- left view
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 230000004927 fusion Effects 0.000 claims abstract description 11
- 238000006243 chemical reaction Methods 0.000 claims abstract description 7
- 230000008859 change Effects 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000009466 transformation Effects 0.000 claims description 5
- 238000013519 translation Methods 0.000 claims description 3
- 238000013461 design Methods 0.000 description 5
- 238000012804 iterative process Methods 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000008030 elimination Effects 0.000 description 2
- 238000003379 elimination reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012216 screening Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- HUTDUHSNJYTCAR-UHFFFAOYSA-N ancymidol Chemical compound C1=CC(OC)=CC=C1C(O)(C=1C=NC=NC=1)C1CC1 HUTDUHSNJYTCAR-UHFFFAOYSA-N 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000010307 cell transformation Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4038—Scaling the whole image or part thereof for image mosaicing, i.e. plane images composed of plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
- G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
-
- 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
- G06T2207/10004—Still image; Photographic image
- G06T2207/10012—Stereo images
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
Abstract
The invention discloses a kind of stereo-picture joining methods for eliminating movement ghost image, comprising: acquires two groups of images using binocular camera, and the view for calculating separately two groups of images is poor;The characteristic point of every group of image is extracted, and the characteristic point is described and is matched, screens the matching to make mistake then to obtain accurate characteristic point to set;New feature constraint condition is set to set according to view difference and characteristic point, and global change is carried out to second group of image using the obtained homography conversion for keeping the feature constraint condition optimal;It determines the overlapping region of first group of image Yu transformed second group of image, and finds out the moving object of overlapping region, Weighted Fusion coefficient is designed according to the relative position of moving object and virtual splicing line;Anastomosing and splicing left view and right view respectively, then spliced left view and right view are synthesized, obtain final perspective view.Stereo-picture joining method of the invention realizes the splicing of high quality for the stereo-picture containing moving object.
Description
Technical field
The present invention relates to computer vision technique and field of image processing more particularly to a kind of solids for eliminating movement ghost image
Image split-joint method.
Background technique
Image mosaic technology is using very extensive, and performance is focused in terms of education, medical treatment, aerospace, amusement
It acts on, is greatly enriched people's lives.Especially as the development of VR, AR and 1,000,000,000 pixels, people are no longer content with having
The image for limiting visual angle starts to pursue higher resolution even the high quality wide viewing angle image of 360 degree of panoramas;This gives traditional monocular figure
As splicing and stereo-picture splicing bring new challenge.For stereo-picture splicing, traditional monocular is not only considered
The problems such as registration and fuzzy ghost image in image mosaic, it is also necessary to obtain having according to parallax (depth) information of stereo-picture vertical
The high quality spliced map of body effect brings viewer one good comfortable 3D experience.
In stereo-picture splicing, there are factors to will affect the perception of last spliced map, including projection distortion, abnormal
Change, horizontal and vertical parallax, ghost etc..And ghost is divided into two classes in splicing result figure, one kind is due to occurring in stitching image
Moving object bring moves ghost, and another kind of is due to the inadequate bring registration ghost of registration accuracy.For there is moving object
The stereo-picture of body splices, and needs to improve splicing precision while being eliminated as much as movement ghost and reduces corresponding distorton
And parallax;And it temporarily can be realized without preferable method splicing essence also can be improved while removal moves ghost in the prior art
It spends and reduces corresponding distorton and parallax.
The disclosure of background above technology contents is only used for auxiliary and understands design and technical solution of the invention, not necessarily
The prior art for belonging to present patent application, no tangible proof show above content present patent application the applying date
In disclosed situation, above-mentioned background technique should not be taken to the novelty and creativeness of evaluation the application.
Summary of the invention
In order to solve the above technical problems, the present invention proposes a kind of stereo-picture joining method for eliminating movement ghost image, for
Stereo-picture containing moving object realizes the splicing of high quality.
In order to achieve the above object, the invention adopts the following technical scheme:
The invention discloses a kind of stereo-picture joining methods for eliminating movement ghost image, comprising the following steps:
S1: acquiring two groups of images using binocular camera, wherein first group of image includes that the first width left view and the first width are right
View, second group of image includes the second width left view and the second width right view, and calculates separately the first width left view and the first width
View difference between right view and the view between the second width left view and the second width right view are poor;
S2: extracting the characteristic point of every group of image, and the characteristic point be described and matched, and then screening makes mistake
Matching is to obtain accurate characteristic point to set;
S3: according in step S1 view difference and step S2 in matched characteristic point new feature constraint is set to set
Condition obtains the homography conversion for keeping the feature constraint condition optimal, is carried out using the homography conversion to second group of image
Global change;
S4: the second group of image converted according to step S3 determines the weight of first group of image Yu transformed second group of image
Folded region, and the moving object of overlapping region is found out, melted according to moving object and the relative position design weighting of virtual splicing line
Collaboration number;
S5: the Weighted Fusion coefficient calculated according to step S4 merges the first width left view and transformed second width respectively
Left view and the first width right view and transformed second width right view, obtain spliced left view and right view, then right
Spliced left view and right view are synthesized, and final perspective view is obtained.
Preferably, step S3 is after carrying out global change to second group of image further include: establishes whole constraint condition Ep,
Content holding is carried out to second group of image of transformation.
Preferably, step S4 is specifically included:
S41: calculate separately the first width left view and transformed second width left view overlapping region error image and
The error image of first width right view and transformed second width right view overlapping region;
S42: binaryzation and hole-filling are carried out to the error image in step S41, determine the region of moving object;
S43: melted according to the region of moving object determined step S42 and the relative position of virtual splicing line to design weighting
Collaboration number.
Preferably, in step S41 the first width left view and transformed second width left view overlapping region error image
Diff1, the first width right view and the error image Diff2 calculation formula of transformed second width right view overlapping region are distinguished
Are as follows:
Wherein,Indicate the grayscale image of the first width left view,Indicate the grayscale image of transformed second width left view,
Indicate the grayscale image of the first width right view,Indicate the grayscale image of transformed second width right view.
Preferably, in step S42 the first width left view and transformed second width left view overlapping region error image
The binarization of Diff1 are as follows:
Wherein, T1 is the binarization threshold of Diff1;
The binarization of first width right view and the error image Diff2 of transformed second width right view overlapping region
Are as follows:
Wherein, T2 is the binarization threshold of Diff2.
Preferably, wherein the value of T1 and T2 respectively between 100~150.
Preferably, the calculation formula of Weighted Fusion coefficient is respectively as follows: in step S43
Wherein, x is the abscissa of any pixel in image overlapping region, xmaxAnd xminRespectively overlapping region right margin is horizontal
Coordinate maximum value and left margin abscissa minimum value, υ are the units that virtual splicing line is translated relative to moving object.
Preferably, when moving object is completely at the left side or right side of the horizontal perpendicular bisector in overlapping region, then virtual splicing
The position of line is the position where the horizontal perpendicular bisector, at this time υ=0;When moving object perpendicular bisector horizontal across overlapping region,
Then the position of virtual splicing line translates υ unit relative to moving object.
Preferably, spliced left view and right view are respectively obtained using following formula in step S5:
Wherein,First width left view respectively to be spliced and transformed second width to be spliced
Left view, IL(x, y) is spliced left view;First width right view respectively to be spliced and wait spell
The transformed second width right view connect, IR(x, y) is spliced right view.
Compared with prior art, the beneficial effects of the present invention are: it is proposed by the present invention eliminate movement ghost image perspective view
As joining method, seamless spliced, reduction ghost image not only may be implemented, but also be capable of detecting when moving object, and eliminate by moving
Ghost image is moved caused by object, so that subsequent splicing is more accurate;It is also improved while removal moves ghost to realize
Splicing precision simultaneously reduces corresponding distorton and parallax.
Detailed description of the invention
Fig. 1 is the flow diagram of the stereo-picture joining method of the elimination movement ghost image of the preferred embodiment of the present invention.
Specific embodiment
Below against attached drawing and in conjunction with preferred embodiment, the invention will be further described.
Such as Fig. 1, the preferred embodiment of the present invention proposes a kind of stereo-picture joining method for eliminating movement ghost image, including with
Lower step:
S1: using binocular camera acquire two groups of images, wherein every group of image include the collected left view of left camera and
The collected right view of right camera, and calculate the disparity map of left and right view between every group of image;
Specifically, collected two groups of images are denoted as I1And I2, wherein every group of image includes the collected left side of left camera
View (With) and the collected right view of right camera (With), and the disparity map of left and right view in every group of image is calculated,
It is denoted as D respectively1And D2。
S2: the characteristic point of every group of image is extracted using feature extraction algorithm (such as SIFT, SURF, ORB), and to characteristic point
It is described and matches, filter out erroneous matching, obtain the set of correct matched characteristic point pair;
Specifically, the characteristic point that every group of image can be extracted using SIFT algorithm (or SURF, ORB scheduling algorithm), is generated
Description of 128 dimensions, calculates arest neighbors and the ratio of time neighbour using nearest neighbor algorithm, is then considered within threshold range
The characteristic point pair matched recycles RANSAC to filter out erroneous matching, obtains correct characteristic point to setWithWhereinFor the first width left view and
The characteristic point of two width left views to set,It is the characteristic point of the first width right view and the second width right view to set,It is the characteristic point of the second width left view and the first width right view to set,For the second width right view and
The characteristic point of one width left view to set,It is the characteristic point of the second width left view and the second width right view to set.
S3: specific feature is established about to set according to matched characteristic point in the disparity map and step S2 in step S1
Beam condition Ef, obtain the homography conversion H for keeping feature constraint condition optimalgGlobal change is carried out, for the viewpoint for keeping general image
Consistency establishes whole constraint condition Ep, content holding is carried out to changing image;
Specifically, it may be implemented by step S3 seamless spliced, comprising:
S31: using the characteristic point of step S2 screening to set
WithAccording to the disparity map D in step S11And D2, design new feature constraint condition Ef:
Ef=γlEl+γrEr+El_r
Wherein γlAnd γrIt is binary number, when splicing two width left views, γlValue is 1, is otherwise 0, same to splice
When two width right views, γrValue is 1, is otherwise 0.Feature constraint condition is made of three parts, left view constraint condition El, right view
Constraint condition ErAnd left and right view constraint condition El_r。
Wherein, left view constraint condition is as follows:
In formula, n1It is characteristic point to setThe number of middle characteristic point, n2It is characteristic point to set Middle spy
Levy the number of point, HlIt is the homography matrix of left view in iterative process;wmAnd wkIt is weighted value, is arrived with according to current signature point
The Gauss distance of all characteristic points is related on the image;
Wherein,Indicate the weighted value of m-th of characteristic point in the second width left eye figure;Indicate the weighted value of k-th of characteristic point in the second width left eye figure.
Similarly available right view constraint condition ErExpression formula:
In formula, n3It is characteristic point to setThe number of middle characteristic point, n4It is characteristic point to setIn
The number of characteristic point, HrIt is the homography matrix of right view in iterative process;wiAnd wjIt is weighted value, and according to current signature point
The Gauss distance of all characteristic points is related on to the image;
Wherein,Indicate the weighted value of ith feature point in the second width right view;Indicate the weighted value of j-th of characteristic point in the second width right view.
Left and right view constraint condition El_rExpression formula is as follows:
In formula, n5It is characteristic point to setThe number of middle characteristic point, Hl_rIt is left or right view in iterative process
Homography matrix;wsIt is weighted value, it is related to according to the Gauss distance of all characteristic points on current signature point to the image;Indicate the weighted value of s-th of characteristic point in the second width left view;
Optimal homography conversion H is iterated to calculate out by these constraint conditionsg, according to HgRespectively by the second width left view
Under the coordinate system for transforming to the first width left view and the first width right view with the second width right view;
S32: for the viewpoint consistency for ensuring two images after converting, whole constraint condition E is establishedp, in image progress
Hold and keep transformation:
Ep=α Eg+βEs+Ey+Ed
In formula, EgRepresent global registration item, EsIt represents shape and retains item, EyRepresent vertical parallax limit entry, EdRepresent level
Parallax limit entry, α, β are weight terms, and it is 0~1 that α, β, which distinguish value,;
Global registration item EgIt embodies as follows:
In formula,WithSet of characteristic points after representation transformation, indicate transformed characteristic point with reference to figure (the
One width figure) in the position of characteristic point should be as consistent as possible;
Shape retains item EsEmbody it is as follows:
In formula,It is transformed three vertex of grid cell, ω respectivelyiThe conspicuousness of expression grid, u=0,Wherein vi、vj、vkIt is three vertex before grid cell transformation respectively,
Vertical parallax limit entry EyIt embodies as follows:
In formula,WithThe y-coordinate of the first width left view and the first width right view is respectively indicated,RespectivelyTable respectively
Show the y-coordinate of the second width left view and the second width right view after converting;
Horizontal parallax limit entry EdIt embodies as follows:
In formula,WithThe x coordinate for converting preceding second width left view and converting preceding second width right view is respectively indicated,WithThe x coordinate of second width right view after respectively indicating the second width left view after converting and converting.
S4: the image converted according to step S3 determines the overlapping region of two images, finds out the moving object of overlapping region
Body, according to moving object virtual splicing line Position Design Weighted Fusion coefficient;
The moving object in binocular image is capable of detecting when by step S4, eliminates the movement ghost image in splicing, tool
Body includes:
S41: after converting left and right view according to step S3, the first width left view and transformed second width left view are calculated separately
The differential chart of the error image Diff1 of figure overlapping region, the first width right view and transformed second width right view overlapping region
As Diff2, calculate as follows:
Wherein,WithIndicate the grayscale image of the first width and transformed second width left view,WithIndicate the first width
With the grayscale image of transformed second width right view.
S42: binaryzation and hole-filling are carried out to the error image in step S41, determine the region of moving object;
The binarization of Diff1 is as follows:
Wherein, T1 be Diff1 binarization threshold, in an experiment T1 can value between 100-150;
Similarly, the binarization of Diff2 are as follows:
Wherein, T2 be Diff2 binarization threshold, in an experiment T2 can value between 100-150;
In some embodiments, above-mentioned T1 and T2 can take same value.
S43: designing new Weighted Fusion coefficient, in conjunction with virtual splicing line, and according to the step S42 moving object determined
The position of moving object under different situations is merged in region, eliminates movement ghost image;
Wherein, the calculation method of weighting coefficient is as follows:
Wherein, x is the abscissa of any pixel in image overlapping region, xmaxAnd xminRespectively overlapping region right margin is horizontal
Coordinate maximum value and left margin abscissa minimum value, υ are the units that virtual splicing line is translated relative to moving object.
When moving object is completely at the left side or right side of the horizontal perpendicular bisector in overlapping region, then the position of virtual splicing line
For the position where the horizontal perpendicular bisector, υ=0 at this time;
When moving object perpendicular bisector horizontal across overlapping region, then the position of virtual splicing line is suitable relative to moving object
Locality translates υ unit (moving object is completely in the left or right side of virtual splicing line after υ unit of translation), when υ is negative,
Then for left, when υ is timing, then for right translation;
Wherein above-mentioned steps S41 and S42 has substantially oriented the region where moving object, and step S43 is according to the positioning
It is merged to calculate Weighted Fusion coefficient further to treat stitching image in the region where moving object out.
S5: according to the Weighted Fusion coefficient of step S4, the first width left view and second transformed left view are merged respectively
Figure, the first width right view and second transformed right view, obtain spliced left and right view ILAnd IR, then left and right is spliced
Figure is synthesized, and final perspective view is obtained.
Wherein, splice left and right view ILAnd IRFormula are as follows:
Wherein,First width left view respectively to be spliced and transformed second width to be spliced
Left view, IL(x, y) is to splice fused left view;First width right view respectively to be spliced and
Transformed second width right view to be spliced, IR(x, y) is to splice fused right view.
The preferred embodiment of the present invention elimination movement ghost image stereo-picture joining method, not only realize it is seamless spliced,
Ghost image is reduced, and is capable of detecting when moving object, and eliminate and move ghost image as caused by moving object, so that subsequent splicing
It is more accurate;To realize removal move ghost while also improve splicing precision and reduce corresponding distorton and
Parallax.
The above content is a further detailed description of the present invention in conjunction with specific preferred embodiments, and it cannot be said that
Specific implementation of the invention is only limited to these instructions.For those skilled in the art to which the present invention belongs, it is not taking off
Under the premise of from present inventive concept, several equivalent substitute or obvious modifications can also be made, and performance or use is identical, all answered
When being considered as belonging to protection scope of the present invention.
Claims (9)
1. a kind of stereo-picture joining method for eliminating movement ghost image, which comprises the following steps:
S1: acquiring two groups of images using binocular camera, wherein first group of image includes the first width left view and the first width right view,
Second group of image includes the second width left view and the second width right view, and calculates separately the first width left view and the first width right view
Between view difference and view between the second width left view and the second width right view it is poor;
S2: the characteristic point of every group of image is extracted, and the characteristic point is described and is matched, then screens the matching to make mistake
To obtain accurate characteristic point to set;
S3: according in step S1 view difference and step S2 in matched characteristic point new feature constraint item is set to set
Part obtains the homography conversion for keeping the feature constraint condition optimal, is carried out using the homography conversion to second group of image complete
Office's transformation;
S4: the second group of image converted according to step S3 determines the overlay region of first group of image Yu transformed second group of image
Domain, and the moving object of overlapping region is found out, Weighted Fusion system is designed according to the relative position of moving object and virtual splicing line
Number;
S5: the Weighted Fusion coefficient calculated according to step S4 merges the first width left view and transformed second width left view respectively
Figure and the first width right view and transformed second width right view, obtain spliced left view and right view, then to splicing
Left view and right view afterwards is synthesized, and final perspective view is obtained.
2. stereo-picture joining method according to claim 1, which is characterized in that step S3 is carried out to second group of image
After global change further include: establish whole constraint condition Ep, content holding is carried out to second group of image of transformation.
3. stereo-picture joining method according to claim 1, which is characterized in that step S4 is specifically included:
S41: the error image and first of the first width left view Yu transformed second width left view overlapping region is calculated separately
The error image of width right view and transformed second width right view overlapping region;
S42: binaryzation and hole-filling are carried out to the error image in step S41, determine the region of moving object;
S43: Weighted Fusion system is designed according to the region of the step S42 moving object determined and the relative position of virtual splicing line
Number.
4. stereo-picture joining method according to claim 3, which is characterized in that in step S41 the first width left view with
Error image Diff1, the first width right view and the transformed right view of second width of transformed second width left view overlapping region
The error image Diff2 calculation formula of figure overlapping region is respectively as follows:
Wherein,Indicate the grayscale image of the first width left view,Indicate the grayscale image of transformed second width left view,Table
Show the grayscale image of the first width right view,Indicate the grayscale image of transformed second width right view.
5. stereo-picture joining method according to claim 3, which is characterized in that in step S42 the first width left view with
The binarization of the error image Diff1 of transformed second width left view overlapping region are as follows:
Wherein, T1 is the binarization threshold of Diff1;
The binarization of first width right view and the error image Diff2 of transformed second width right view overlapping region are as follows:
Wherein, T2 is the binarization threshold of Diff2.
6. stereo-picture joining method according to claim 5, which is characterized in that wherein the value of T1 and T2 exists respectively
Between 100~150.
7. stereo-picture joining method according to claim 3, which is characterized in that Weighted Fusion coefficient in step S43
Calculation formula is respectively as follows:
Wherein, x is the abscissa of any pixel in image overlapping region, xmaxAnd xminRespectively overlapping region right margin abscissa
Maximum value and left margin abscissa minimum value, υ are the units that virtual splicing line is translated relative to moving object.
8. stereo-picture joining method according to claim 7, which is characterized in that when moving object is completely in overlapping region
When the left side or right side of horizontal perpendicular bisector, then the position of virtual splicing line be the horizontal perpendicular bisector where position, at this time υ=
0;When moving object perpendicular bisector horizontal across overlapping region, then the position of virtual splicing line is relative to moving object translation υ
Unit.
9. stereo-picture joining method according to claim 7, which is characterized in that respectively obtained in step S5 using following formula
Spliced left view and right view:
Wherein,For the first width left view to be spliced,For transformed second width left view to be spliced, IL
(x, y) is spliced left view;For the first width right view to be spliced,It is to be spliced transformed
Second width right view, IR(x, y) is spliced right view.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811389589.3A CN109493282A (en) | 2018-11-21 | 2018-11-21 | A kind of stereo-picture joining method for eliminating movement ghost image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811389589.3A CN109493282A (en) | 2018-11-21 | 2018-11-21 | A kind of stereo-picture joining method for eliminating movement ghost image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109493282A true CN109493282A (en) | 2019-03-19 |
Family
ID=65697262
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811389589.3A Pending CN109493282A (en) | 2018-11-21 | 2018-11-21 | A kind of stereo-picture joining method for eliminating movement ghost image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109493282A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110020985A (en) * | 2019-04-12 | 2019-07-16 | 广西师范大学 | A kind of video-splicing system and method for Binocular robot |
CN110619652A (en) * | 2019-08-19 | 2019-12-27 | 浙江大学 | Image registration ghost elimination method based on optical flow mapping repeated area detection |
CN111311750A (en) * | 2020-01-17 | 2020-06-19 | 武汉大学 | Mosaic line network global optimization method based on constrained triangulation network |
CN112200727A (en) * | 2020-11-06 | 2021-01-08 | 厦门星宸科技有限公司 | Image stitching device, image processing chip, and image stitching method |
CN112907704A (en) * | 2021-02-04 | 2021-06-04 | 浙江大华技术股份有限公司 | Image fusion method, computer equipment and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104134200A (en) * | 2014-06-27 | 2014-11-05 | 河海大学 | Mobile scene image splicing method based on improved weighted fusion |
WO2016086754A1 (en) * | 2014-12-03 | 2016-06-09 | 中国矿业大学 | Large-scale scene video image stitching method |
CN105678687A (en) * | 2015-12-29 | 2016-06-15 | 天津大学 | Stereo image stitching method based on content of images |
CN108470324A (en) * | 2018-03-21 | 2018-08-31 | 深圳市未来媒体技术研究院 | A kind of binocular stereo image joining method of robust |
-
2018
- 2018-11-21 CN CN201811389589.3A patent/CN109493282A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104134200A (en) * | 2014-06-27 | 2014-11-05 | 河海大学 | Mobile scene image splicing method based on improved weighted fusion |
WO2016086754A1 (en) * | 2014-12-03 | 2016-06-09 | 中国矿业大学 | Large-scale scene video image stitching method |
CN105678687A (en) * | 2015-12-29 | 2016-06-15 | 天津大学 | Stereo image stitching method based on content of images |
CN108470324A (en) * | 2018-03-21 | 2018-08-31 | 深圳市未来媒体技术研究院 | A kind of binocular stereo image joining method of robust |
Non-Patent Citations (3)
Title |
---|
徐敏: "改进加权融合在消除图像拼接重影中的运用", 《软件工程》 * |
潘恒辉: "图像拼接中鬼影消除算法研究", 《舰船电子工程》 * |
瞿中等: "一种消除图像拼接缝和鬼影的快速拼接算法", 《计算机科学》 * |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110020985A (en) * | 2019-04-12 | 2019-07-16 | 广西师范大学 | A kind of video-splicing system and method for Binocular robot |
CN110619652A (en) * | 2019-08-19 | 2019-12-27 | 浙江大学 | Image registration ghost elimination method based on optical flow mapping repeated area detection |
CN110619652B (en) * | 2019-08-19 | 2022-03-18 | 浙江大学 | Image registration ghost elimination method based on optical flow mapping repeated area detection |
CN111311750A (en) * | 2020-01-17 | 2020-06-19 | 武汉大学 | Mosaic line network global optimization method based on constrained triangulation network |
CN112200727A (en) * | 2020-11-06 | 2021-01-08 | 厦门星宸科技有限公司 | Image stitching device, image processing chip, and image stitching method |
CN112200727B (en) * | 2020-11-06 | 2023-11-21 | 星宸科技股份有限公司 | Image stitching device, image processing chip, and image stitching method |
CN112907704A (en) * | 2021-02-04 | 2021-06-04 | 浙江大华技术股份有限公司 | Image fusion method, computer equipment and device |
CN112907704B (en) * | 2021-02-04 | 2024-04-12 | 浙江大华技术股份有限公司 | Image fusion method, computer equipment and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109493282A (en) | A kind of stereo-picture joining method for eliminating movement ghost image | |
AU2017246716B2 (en) | Efficient determination of optical flow between images | |
Itoh et al. | Interaction-free calibration for optical see-through head-mounted displays based on 3d eye localization | |
CN109685913B (en) | Augmented reality implementation method based on computer vision positioning | |
CN106504194B (en) | A kind of image split-joint method based on best splicing plane and local feature | |
CN112634341B (en) | Method for constructing depth estimation model of multi-vision task cooperation | |
CN108470324A (en) | A kind of binocular stereo image joining method of robust | |
CN110211043A (en) | A kind of method for registering based on grid optimization for Panorama Mosaic | |
CN107767339B (en) | Binocular stereo image splicing method | |
CN110956661B (en) | Method for calculating dynamic pose of visible light and infrared camera based on bidirectional homography matrix | |
CN106469444B (en) | Rapid image fusion method for eliminating splicing gap | |
CN110111248A (en) | A kind of image split-joint method based on characteristic point, virtual reality system, camera | |
CN112085659A (en) | Panorama splicing and fusing method and system based on dome camera and storage medium | |
CN105894443B (en) | A kind of real-time video joining method based on improved SURF algorithm | |
CN103945207B (en) | A kind of stereo-picture vertical parallax removing method based on View Synthesis | |
CN110099268B (en) | Blind area perspective display method with natural color matching and natural display area fusion | |
CN112926464B (en) | Face living body detection method and device | |
CN113744315B (en) | Semi-direct vision odometer based on binocular vision | |
CN110120013A (en) | A kind of cloud method and device | |
CN105657401A (en) | Naked eye 3D display method and system and naked eye 3D display device | |
CN106023170A (en) | Binocular 3D distortion rectification method based on GPU | |
CN108133469A (en) | Light field splicing apparatus and method based on EPI | |
CN108830804B (en) | Virtual-real fusion fuzzy consistency processing method based on line spread function standard deviation | |
CN110111255A (en) | A kind of stereo-picture joining method | |
Yang et al. | Depth from water reflection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |