CN105976324B - Vehicle image splicing method - Google Patents
Vehicle image splicing method Download PDFInfo
- Publication number
- CN105976324B CN105976324B CN201610335450.5A CN201610335450A CN105976324B CN 105976324 B CN105976324 B CN 105976324B CN 201610335450 A CN201610335450 A CN 201610335450A CN 105976324 B CN105976324 B CN 105976324B
- Authority
- CN
- China
- Prior art keywords
- image
- target
- feature space
- common feature
- energy function
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 23
- 238000013507 mapping Methods 0.000 claims description 28
- 238000012545 processing Methods 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 32
- 238000001514 detection method Methods 0.000 description 8
- 239000002245 particle Substances 0.000 description 8
- 238000004364 calculation method Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000013519 translation Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000010276 construction Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000001846 repelling effect Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012800 visualization Methods 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
-
- G06T3/18—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/32—Indexing scheme for image data processing or generation, in general involving image mosaicing
Abstract
The invention discloses a vehicle image splicing method including that an original environment image is acquired by a vehicle camera, the splicing area corresponding to the original environment image is queried in a target image, and furthermore, a direction gradient graph is constructed based on the relevance between the original environment image and the splicing area; the direction gradient graph is subjected to regularization processing to obtain a common feature space, and furthermore, an energy function of the common feature space is constructed based on the relevance of the similarity distance measurement and the common feature space; and an optimal solution for enabling the value of the energy function to minimum is obtained and is taken as the final pixel value of the original environment image, and furthermore, the display is carried out after the final pixel value is assigned to the splicing area. According to the vehicle image splicing method, the computational complexity is reduced, the computational speed is improved, and the problem that the distortion rate of the spliced image is lowered because the matching of the image matching feature points is low can be prevented.
Description
Technical field
The invention belongs to technical field of automotive electronics, more particularly to a kind of vehicle-mounted image split-joint method.
Background technology
Panorama Mosaic is developed rapidly as emerging technology in recent years, also obtains more and more researchers
Concern.In Panoptic visualization auxiliary is parked, need to generate the panorama around vehicle body by the technology of Panorama Mosaic
View.
At present, panorama mosaic method of the prior art is come using a global homography matrix with a homography matrix
Represent the perspective transform relation between the image of input.By taking two image mosaics as an example, the matching of two images for obtaining first is special
Levy a little, so-called matching characteristic point is that two characteristic points spatially represent same point;Then according to the matching characteristic point for obtaining, solve
Homography matrix;Finally, all pixels on wherein piece image are clicked through by line translation according to the homography matrix, it is determined that in another width
The correspondence position of image place plane, that is, obtain the splicing result of two width figures, so as to further to image carry out color blend etc.
Process, obtain preferable spliced map.
But, there is computationally intensive, the slow shortcoming of calculating speed in this joining method, and it is special images match easily occur
The matching levied a little is low so that the pattern distortion generation rate of splicing is reduced.
The content of the invention
The technical problem to be solved is, there is provided a kind of vehicle-mounted image split-joint method and system, can reduce
Amount of calculation, improves calculating speed, it is to avoid the matching for images match characteristic point occur is low so that the pattern distortion generation rate of splicing
The problems such as reduction.
To solve above-mentioned technical problem, the embodiment of the present invention provides a kind of vehicle-mounted image split-joint method, and methods described includes:
A, the camera acquisition primal environment image by being arranged at automobile, and inquire in the target image and the original
The corresponding splicing regions of beginning ambient image, and the pass being based further between the primal environment image and the splicing regions
Connection property builds direction gradient figure;
B, Regularization is carried out to the direction gradient figure, obtain common feature space, and be based further on similitude
Relevance between distance measure and the common feature space for obtaining, constructs the energy letter in the common feature space
Number;
C, the optimal solution for solving the value minimum for causing the energy function, using the optimal solution as the primal environment figure
The final pixel value of picture, and further according to the final pixel value be assigned to after the splicing regions show.
Wherein, in step a " splice region corresponding with the primal environment image is inquired in the target image
Realized by inquiring about default target image splicing mapping table in domain ";Wherein, the map information at least includes primal environment figure
The sequence number of picture, the pixel coordinate information of primal environment image.
Wherein, the concrete steps of the default target image splicing mapping table are realized as follows:
Target image is divided into into multiple target areas according to world coordinate system, and will be positioned at two target area intersections
Preset range in region be defined as splicing regions, and determine each target area and the original ring corresponding to splicing regions
The sequence number of border image;
According to the mapping relations between primal environment image coordinate system and world coordinate system, and the coordinate system of target image
Mapping relations between world coordinate system, obtain each target pixel points and original image maps mutually on the target image
Map information;Wherein, in the splicing regions of target image, each target pixel points primal environment figure different from two width respectively
As a upper specific pixel point maps mutually;
By the corresponding mapping letter of the positional information and each target pixel points of each target pixel points of the target image
Breath is preserved, and obtains target image splicing mapping table.
Wherein, step b is specifically included:
By formulaRegularization is carried out to the direction gradient figure
Process, obtain common feature space;Wherein, G (x, y) is the common feature space for obtaining;| ▽ I (x, y) | is the side
To the gradient modulus value of gradient map;W (x, y) is the center window with certain size in the direction gradient figure;K is fixation
Constant, can value be 100;
Based on variation principle, the energy function in the common feature space is constituted using the combination of various distance measures, made
Obtaining the energy function in the common feature space can be expressed as:E (p, x, y)=- EN(p,x,y)+λ1EH(p,x,y)-λ2EG(p,
x,y);Wherein, E (p, x, y) is the energy function in the common feature space;EN(p,x,y),EH(p,x,y),EG(p, x, y) point
Biao Shi not be based on and go the normalizated correlation coefficient of average to measure energy function, the energy function based on Hausdorff distances and base
In the energy function of local maximum mask statistic;λ1, λ2Represent Lagrange multiplier weight;P represents anamorphose parameter.
Implement the embodiment of the present invention, with following beneficial effect:
The vehicle-mounted image split-joint method that the present invention is provided, single pass completes the detection of primal environment image, and in detection
While complete the estimation of target distortion parameter, solve (imagings of the same target under different observation geometry of multiple target in image
Performance) difficult problem such as quick, high detection rate, amount of calculation is reduced, calculating speed is improve, realize the matching of images match characteristic point
The purpose of reliability height and high precision, it is to avoid the problems such as pattern distortion generation rate of splicing is reduced.
Description of the drawings
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, may be used also
To obtain other accompanying drawings according to these accompanying drawings.
Fig. 1 is a kind of schematic flow sheet of vehicle-mounted image split-joint method provided in an embodiment of the present invention;
Fig. 2 is to carry out area to target image in a kind of vehicle-mounted image split-joint method application scenarios provided in an embodiment of the present invention
The structural representation that domain divides;
Fig. 3 is the structural representation of a kind of vehicle-mounted image mosaic system that the present invention provides embodiment.
Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than the embodiment of whole.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
As shown in figure 1, in the embodiment of the present invention, there is provided a kind of vehicle-mounted image split-joint method, methods described includes:
Step S10, by be arranged at automobile camera obtain primal environment image, and inquire in the target image with
The corresponding splicing regions of the primal environment image, and be based further on the primal environment image and the splicing regions it
Between relevance build direction gradient figure;
Specifically, obtaining a primal environment image, each width primal environment by being arranged at any one camera of automobile
Image is to there is the information such as a sequence number, coordinate position, the pixel value of pixel of each width primal environment image to be remembered
Record is got off.Now, the panoramic picture (final spliced aerial view) that target image as needs to splice and show, and splice region
Domain is realized by inquiring about default target image splicing mapping table, and be stored with target figure in target image splicing mapping table
The corresponding relation between sequence number, the pixel coordinate of each pixel and primal environment image as in.
In order to exclude impact of the ambient noise with imaging illumination condition to primal environment image, obtain pixel and more accurately spell
Panorama sketch is connect, therefore devises a kind of feature space of novelty:The feature space of regularization gradient.This feature space is by calculating
Level is realized with the gradient modulus value of vertical direction, it is therefore desirable to based on the relevance between primal environment image and splicing regions
Direction gradient figure is formed, direction gradient map is two-dimensional gradient figure, and it is by horizontal direction gradient map and vertical gradient figure
Into.
The construction of direction gradient figure, can be realized by formula (1):
▽ I (x, y)=▽ Ix(x,y)+i×▽Iy(x,y) (1);
In formula (1), ▽ Ix(x, y), ▽ Iy(x, y) represent respectively x and y to gradient map;Wherein, ▽ Ix(x, y)=I (x+
1, y)+I (x-1, y) -2 × I (x, y), ▽ Iy(x, y)=I (x, y+1)+I (x, y-1) -2 × I (x, y);
Because ▽ I (x, y) is a plural number, then its range value and its gradient direction (gradient direction angle) can be calculated, its meter
Calculate formula (1) deformation as follows:
θ=atg (▽ Iy(x,y)/▽Ix(x,y)) (3);
In embodiments of the present invention, the concrete steps of target image splicing mapping table are realized as follows:
(1) target image is divided into into multiple target areas according to world coordinate system, and will be handed over positioned at two target areas
The region in preset range at boundary is defined as splicing regions, and determines each target area and the original corresponding to splicing regions
The sequence number of beginning ambient image;
For ease of understanding, refer to shown in Fig. 2, it illustrates one embodiment that region division is carried out to target image
Schematic diagram, in this embodiment, in the target image, will go out specific F, B, L, R in automobile region division all around
Four regions, wherein, point P1~P4 for the surrounding of automobile four summits, and the line segment that is made up of point P3, P4 and by point P5, P6
The line segment of composition forms two borders of target image, and by measurement h0~h3 in target image, w1, H_CAR, W_ can be obtained
The concrete range information of CAR etc., wherein, H_CAR is the length of automobile image in target image, and W_CAR is in target image
The width of automobile image, it is to be understood that closed with the scaling translation of the physical length of automobile etc. according to the length of the H_CAR
System, may be used to determine the mapping relations in pixel and primal environment image between the coordinate of pixel in target image.Together
When, it may be determined that the pixel value of the pixel in each region is from which width primal environment image, for example, the pixel in F regions
Pixel value can come from camera before automobile (camera is represented in Fig. 3 with the constitutional diagram of small circle and little square frame,
Primal environment image captured by similarly hereinafter), the pixel value of the pixel in Zone R domain is from original captured by car right side camera
Ambient image, the pixel value of the pixel in L regions from the primal environment image captured by automobile left side camera, B regions
The pixel value of pixel is from the primal environment image captured by automotive back camera;Original ring captured by different cameras
Border image can be made a distinction by different ambient image sequence numbers i.
(2) according to the mapping relations between primal environment image coordinate system and world coordinate system, and the seat of target image
Mapping relations between mark system and world coordinate system, obtain each target pixel points and original image maps mutually on target image
Map information;Wherein, in the splicing regions of target image, each target pixel points primal environment figure different from two width respectively
As a upper specific pixel point maps mutually;
It is understood that in the splicing regions of target image, the fusion due to needing to carry out pixel, each pixel
A specific pixel point maps mutually on point primal environment image different from two width respectively;For example, A1 regions in target image
Pixel pixel value, need in picture according to captured by automobile front side camera and left side camera respectively one it is specific
The pixel of the pixel of point carries out calculating acquisition;
Specifically, remember that primal environment image coordinate is (u0i,v0i), target image coordinate be (u, v), the target image
Upper each pixel is primal environment image coordinate (u with the mapping relations of original image0i,v0i) and ambient image sequence number i
With the mapping relations between target image coordinate (u, v).Take world coordinates (xw,yw,zw) as intermediate quantity, find out (u, v) respectively
With (u0i,v0i) the two is with its mapping relations, it is hereby achieved that (u, v) and (u0i,v0i) between mapping relations.
Specifically, in one example, the coordinate mapping relations of primal environment image determine by imaging model, for example, take the photograph
As the relation between head coordinate system and coordinates of original image coordinates system, such as scara models can be adopted.
And mapping relations of the target image (top view) and world coordinates between are relatively simple, world coordinates to top view is sat
Only through scaling, the conversion process of translation between mark.In short, needing the region for showing to be around target image and body of a motor car
Scaling relation, therefore by the coordinate of target image, corresponding vehicle body coordinate can be calculated, then joined by outside camera, calculate corresponding
Camera coordinate, finally according to camera internal reference, calculate the coordinate of primal environment image.
During above-mentioned Coordinate Conversion, from (u, v) world coordinates (x is mapped tow,yw,zw) after, you can according to (xw,
yw,zw) belonging to region (F, L, R, B) determining the value of sequence number i of ambient image.
(3) by the positional information of each target pixel points of target image and the corresponding map information of each target pixel points
Preserved, obtained target image splicing mapping table.
Step S20, Regularization is carried out to the direction gradient figure, obtain common feature space, and be based further on
Similarity distance is estimated and the relevance between the common feature space for obtaining, and constructs the energy in the common feature space
Flow function;
Specifically, carrying out Regularization to direction gradient map by formula (4), common feature space is obtained, specifically such as
Under;
In formula (4), G (x, y) is the common feature space for obtaining;| ▽ I (x, y) | for the direction gradient figure ladder
Degree modulus value;W (x, y) is the center window with certain size in the direction gradient figure;K is fixed constant, can value
For 100;
Based on variation principle, the energy function in common feature space is constituted using the combination of various distance measures so that altogether
Property feature space energy function by formula (5) represent, it is specific as follows:
E (p, x, y)=- EN(p,x,y)+λ1EH(p,x,y)-λ2EG(p,x,y) (5);
In formula (5), E (p, x, y) is the energy function in common feature space;EN(p,x,y),EH(p,x,y),EG(p,x,y)
Respectively represent based on go average normalizated correlation coefficient measure energy function, the energy function based on Hausdorff distances and
Energy function based on local maximum mask statistic;λ1, λ2Represent Lagrange multiplier weight;P represents anamorphose parameter.
Under the conditions of affine, the distorted pattern such as formula (6) of image:
At this moment p=(a, b, c, d, Δ x, Δ y)T。
When certain conditions are met, the distorted pattern formula (6) of image can be reduced to formula (7):
Wherein p=(a, b, c, d, Δ x, Δ y)T。
In summary, the acquisition of anamorphose parameter P, then represent the confirmation of desired value in common feature space so that every
One anamorphose parameter P is represented as a pixel value, therefore can pass through each on the energy function in common feature space
The quick positioning (i.e. each Local Extremum is corresponding to a detection target) of individual Local Extremum, so as to realize primal environment
Matching between image and target image.
Step S30, the optimal solution for causing the value of the energy function minimum is solved, using the optimal solution as described original
The final pixel value of ambient image, and further according to the final pixel value be assigned to after the splicing regions show.
Specifically, being optimized to the energy function in common feature space using multiparticle group algorithm, common feature is tried to achieve
The optimal value of anamorphose parameter P in the energy function in space, using optimal solution as primal environment image final pixel value, and
Further it is assigned to be shown after splicing regions according to final pixel value, so as to possess the incomparable effect of traditional algorithm and excellent
Gesture;Simultaneously in order to reach the arithmetic result of more accurate and efficiency, and population can be dynamically generated or eliminate, reduction need not
The amount of calculation wanted, so as to reach the purpose of the reliability for realizing images match height and high precision, adds convergence and repellency
Judgement, increased the interaction between population.
The energy function in common feature space using multiparticle group's algorithm to implement step as follows:
Step S301, the parameter for determining population;Wherein, parameter includes the primary equal with the quantity of matching detection
Group's quantity, and also include that population convergence radius, population are repelled radius, individual Studying factors, population Studying factors and be used to
Property weight;
Step S302, initialization population, it includes that it is 0, Yi Jisui to arrange maximum iteration time and primary iteration number of times
Machine sets the speed and its corresponding velocity attitude of each population, and the space bit of each particle in each population is set at random
Put;
Step S303, acquisition current iteration number of times, and whether the current iteration number of times for judging to get is less than greatest iteration
Number of times;If it is, performing next step S304;If it is not, then redirecting execution step S305;
Step S304, the current iteration number of times for getting plus one, and each particle in each population is traveled through more
Newly, and according to the default individual Studying factors, population Studying factors and inertia weight, and according to grain in single particle group
Optimum of the particle in current traversal evolution track in optimal value and all populations of the son in current traversal evolution track
Value, obtains traveling through each particle group velocity after updating;And
According to each particle group velocity after the traversal renewal for obtaining, the position of each population is updated, and it is sequentially right
Population after more new position carries out repelling after determination processing and convergence determination processing, return to step S303;
Step S305, termination carry out traversal renewal to each particle in each population, and filter out grain in each population
Optimal value of the son in the previous evolution track for terminating traversal, and further by the optimal value work of particle in each population of screening
For the output of corresponding Local Extremum.
As shown in figure 3, in the embodiment of the present invention, there is provided a kind of vehicle-mounted image mosaic system, the system includes:
Direction gradient figure construction unit 10, for the camera acquisition primal environment image by being arranged at automobile, and
The splicing regions corresponding with the primal environment image are inquired in target image, and is based further on the primal environment figure
Relevance between picture and the splicing regions builds direction gradient figure;
Energy function acquiring unit 12, for carrying out Regularization to the direction gradient figure, obtains common feature empty
Between, and be based further on similarity distance and estimate and the relevance between the common feature space for obtaining, construct described
The energy function in common feature space;
Splicing unit 14, the optimal solution minimum for solving the value for causing the energy function, by the optimal solution
As the final pixel value of the primal environment image, and further the splicing regions are assigned to according to the final pixel value
After show.
Wherein, the splicing regions are realized by inquiring about default target image splicing mapping table;Wherein, the mapping
Information is at least including sequence number, the pixel coordinate information of primal environment image of primal environment image.
Wherein, the energy function acquiring unit 12 includes:
Common feature space acquisition module, for by formulaIt is right
The direction gradient figure carries out Regularization, obtains common feature space;Wherein, G (x, y) is the common feature for obtaining
Space;| ▽ I (x, y) | for the direction gradient figure gradient modulus value;W (x, y) is have certain face in the direction gradient figure
The center window of product size;K is fixed constant, can value be 100;
Energy function acquisition module, for based on variation principle, using the combination of various distance measures the general character being constituted
The energy function of feature space so that the energy function in the common feature space can be expressed as:E (p, x, y)=- EN(p,
x,y)+λ1EH(p,x,y)-λ2EG(p,x,y);Wherein, E (p, x, y) is the energy function in the common feature space;EN(p,x,
y),EH(p,x,y),EG(p, x, y) is represented respectively and is measured energy function, is based on based on the normalizated correlation coefficient for going average
The energy function of Hausdorff distances and the energy function based on local maximum mask statistic;λ1, λ2Represent Lagrange
Multiplier weight;P represents anamorphose parameter.
Implement the embodiment of the present invention, with following beneficial effect:
The vehicle-mounted image split-joint method that the present invention is provided, single pass completes the detection of primal environment image, and in detection
While complete the estimation of target distortion parameter, solve (imagings of the same target under different observation geometry of multiple target in image
Performance) difficult problem such as quick, high detection rate, amount of calculation is reduced, calculating speed is improve, realize the matching of images match characteristic point
The purpose of reliability height and high precision, it is to avoid the problems such as pattern distortion generation rate of splicing is reduced.
It should be noted that in said system embodiment, each included system unit simply enters according to function logic
What row was divided, but above-mentioned division is not limited to, as long as corresponding function can be realized;In addition, each functional unit
Specific name is also only to facilitate mutually differentiation, is not limited to protection scope of the present invention.
One of ordinary skill in the art will appreciate that realizing that all or part of step in above-described embodiment method can be
Related hardware is instructed to complete by program, described program can be stored in a computer read/write memory medium,
Described storage medium, such as ROM/RAM, disk, CD.
Above disclosed is only a kind of preferred embodiment of the invention, can not limit the power of the present invention with this certainly
Sharp scope, therefore the equivalent variations made according to the claims in the present invention, still belong to the scope that the present invention is covered.
Claims (2)
1. a kind of vehicle-mounted image split-joint method, it is characterised in that methods described includes:
A, the camera acquisition primal environment image by being arranged at automobile, and inquire in the target image and the original ring
The corresponding splicing regions of border image, and the relevance being based further between the primal environment image and the splicing regions
Build direction gradient figure;
B, Regularization is carried out to the direction gradient figure, obtain common feature space, and be based further on similarity distance
Estimate and the relevance between the common feature space for obtaining, construct the energy function in the common feature space;
C, the optimal solution for solving the value minimum for causing the energy function, using the optimal solution as the primal environment image
Final pixel value, and further according to the final pixel value be assigned to after the splicing regions show;
" splicing regions corresponding with the primal environment image are inquired in the target image " in step a by inquiry
Default target image splices mapping table to realize;Wherein, the map information of the mapping table at least includes primal environment image
Sequence number, the pixel coordinate information of primal environment image;
The concrete steps of the default target image splicing mapping table are realized as follows:
Target image is divided into into multiple target areas according to world coordinate system, and by positioned at the pre- of two target area intersections
Determine the region in scope and be defined as splicing regions, and determine each target area and the primal environment figure corresponding to splicing regions
The sequence number of picture;
According to the mapping relations between primal environment image coordinate system and world coordinate system, and the coordinate system and generation of target image
Mapping relations between boundary's coordinate system, obtain the mapping of each target pixel points and original image maps mutually on the target image
Information;Wherein, in the splicing regions of target image, on each target pixel points primal environment image different from two width respectively
A specific pixel point maps mutually;
The positional information of each target pixel points of the target image and the corresponding map information of each target pixel points are entered
Row is preserved, and obtains target image splicing mapping table.
2. the method for claim 1, it is characterised in that step b is specifically included:
By formulaRegularization is carried out to the direction gradient figure,
Obtain common feature space;Wherein, G (x, y) is the common feature space for obtaining;For the direction gradient figure
Gradient modulus value;W (x, y) is the center window with certain size in the direction gradient figure;K is fixed constant, can
Value is 100;
Based on variation principle, the energy function in the common feature space is constituted using the combination of various distance measures so that institute
Stating the energy function in common feature space can be expressed as:E (p, x, y)=- EN(p,x,y)+λ1EH(p,x,y)-λ2EG(p,x,
y);Wherein, E (p, x, y) is the energy function in the common feature space;EN(p,x,y),EH(p,x,y),EG(p, x, y) difference
Represent and measure energy function, the energy function based on Hausdorff distances based on the normalizated correlation coefficient for going average and be based on
The energy function of local maximum mask statistic;λ1, λ2Represent Lagrange multiplier weight;P represents anamorphose parameter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610335450.5A CN105976324B (en) | 2016-05-19 | 2016-05-19 | Vehicle image splicing method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610335450.5A CN105976324B (en) | 2016-05-19 | 2016-05-19 | Vehicle image splicing method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105976324A CN105976324A (en) | 2016-09-28 |
CN105976324B true CN105976324B (en) | 2017-05-17 |
Family
ID=56957109
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610335450.5A Expired - Fee Related CN105976324B (en) | 2016-05-19 | 2016-05-19 | Vehicle image splicing method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105976324B (en) |
Families Citing this family (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106780344B (en) * | 2017-01-06 | 2020-07-07 | 苏州大学 | Automobile 360-degree all-round-view image splicing method based on deviation transfer and splitting |
CN107194878A (en) * | 2017-06-21 | 2017-09-22 | 微鲸科技有限公司 | Image split-joint method and device |
CN108986183B (en) * | 2018-07-18 | 2022-12-27 | 合肥亿图网络科技有限公司 | Method for manufacturing panoramic map |
CN110827197A (en) * | 2019-10-08 | 2020-02-21 | 武汉极目智能技术有限公司 | Method and device for detecting and identifying vehicle all-round looking target based on deep learning |
CN113538237A (en) * | 2021-07-09 | 2021-10-22 | 北京超星未来科技有限公司 | Image splicing system and method and electronic equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102201115A (en) * | 2011-04-07 | 2011-09-28 | 湖南天幕智能科技有限公司 | Real-time panoramic image stitching method of aerial videos shot by unmanned plane |
CN103136720A (en) * | 2013-03-12 | 2013-06-05 | 中科院微电子研究所昆山分所 | Vehicle-mounted 360-degree panorama mosaic method |
CN103985133A (en) * | 2014-05-30 | 2014-08-13 | 武汉大学 | Search method and system for optimal splicing lines among images based on graph-cut energy optimization |
-
2016
- 2016-05-19 CN CN201610335450.5A patent/CN105976324B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102201115A (en) * | 2011-04-07 | 2011-09-28 | 湖南天幕智能科技有限公司 | Real-time panoramic image stitching method of aerial videos shot by unmanned plane |
CN103136720A (en) * | 2013-03-12 | 2013-06-05 | 中科院微电子研究所昆山分所 | Vehicle-mounted 360-degree panorama mosaic method |
CN103985133A (en) * | 2014-05-30 | 2014-08-13 | 武汉大学 | Search method and system for optimal splicing lines among images based on graph-cut energy optimization |
Also Published As
Publication number | Publication date |
---|---|
CN105976324A (en) | 2016-09-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105976324B (en) | Vehicle image splicing method | |
CN110569704B (en) | Multi-strategy self-adaptive lane line detection method based on stereoscopic vision | |
US10268201B2 (en) | Vehicle automated parking system and method | |
US11350073B2 (en) | Disparity image stitching and visualization method based on multiple pairs of binocular cameras | |
US20220044356A1 (en) | Large-field-angle image real-time stitching method based on calibration | |
JP5588812B2 (en) | Image processing apparatus and imaging apparatus using the same | |
CN110842940A (en) | Building surveying robot multi-sensor fusion three-dimensional modeling method and system | |
CN112037159B (en) | Cross-camera road space fusion and vehicle target detection tracking method and system | |
CN106856000A (en) | A kind of vehicle-mounted panoramic image seamless splicing processing method and system | |
CN111932627B (en) | Marker drawing method and system | |
CN111768332A (en) | Splicing method of vehicle-mounted all-around real-time 3D panoramic image and image acquisition device | |
JP2023505891A (en) | Methods for measuring environmental topography | |
CN114693787A (en) | Parking garage map building and positioning method and system and vehicle | |
CN112308928A (en) | Camera without calibration device and laser radar automatic calibration method | |
TWI682361B (en) | Method and system for road image reconstruction and vehicle positioning | |
CN116912715A (en) | Unmanned aerial vehicle vision servo control method and system for fan blade inspection | |
CN109490926B (en) | Path planning method based on binocular camera and GNSS | |
CN105957010A (en) | Vehicle-mounted image splicing system | |
CN114339185A (en) | Image colorization for vehicle camera images | |
CN111476798B (en) | Vehicle space morphology recognition method and system based on contour constraint | |
CN112819711A (en) | Monocular vision-based vehicle reverse positioning method utilizing road lane line | |
Lin et al. | Topview transform model for the vehicle parking assistance system | |
CN103236053B (en) | A kind of MOF method of moving object detection under mobile platform | |
CN104021535B (en) | The method of stepping framing ccd image splicing | |
CN115239822A (en) | Real-time visual identification and positioning method and system for multi-module space of split type flying vehicle |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170517 |