CN106856000A - A kind of vehicle-mounted panoramic image seamless splicing processing method and system - Google Patents
A kind of vehicle-mounted panoramic image seamless splicing processing method and system Download PDFInfo
- Publication number
- CN106856000A CN106856000A CN201510903646.5A CN201510903646A CN106856000A CN 106856000 A CN106856000 A CN 106856000A CN 201510903646 A CN201510903646 A CN 201510903646A CN 106856000 A CN106856000 A CN 106856000A
- Authority
- CN
- China
- Prior art keywords
- target
- image
- pixel
- pixel points
- splicing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 238000013507 mapping Methods 0.000 claims abstract description 59
- 238000000034 method Methods 0.000 claims description 27
- 238000012545 processing Methods 0.000 claims description 13
- 230000000149 penetrating effect Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 7
- 238000005516 engineering process Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007499 fusion processing Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000013519 translation Methods 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
- 238000013459 approach Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000009466 transformation Effects 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 transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4038—Image mosaicing, e.g. composing plane images from plane sub-images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- 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
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of vehicle-mounted panoramic image seamless splicing processing method, comprise the following steps：Primal environment image is obtained by the camera for being arranged at automobile surrounding respectively；Mapping table is spliced by a target image for previously generating, the map information corresponding to each target pixel points in acquisition target image is inquired about, wherein, the map information at least sequence number, pixel coordinate information including primal environment image；According to the map information of each target pixel points in the target image, obtain the pixel value on respective pixel coordinate on the primal environment image of corresponding sequence number, and the final pixel value of each target pixel points on target image is obtained according to the pixel value, and the final pixel value is assigned to corresponding target pixel points.Implement the present invention, the speed and splicing effect of Panorama Mosaic can be improved.
Description
Technical field
The invention belongs to technical field of automotive electronics, more particularly to a kind of vehicle-mounted panoramic for auxiliary of parking
Image seamless splicing processing method and system.
Background technology
Panorama Mosaic is developed rapidly as emerging technology in recent years, also obtains more and more
The concern of researcher., it is necessary to pass through the technology next life of Panorama Mosaic in Panoptic visualization auxiliary is parked
Into the panoramic view around vehicle body.Wherein, seamless spliced technology is the core of panorama system, and image seamless is spelled
It is image by a series of overlaps in space of aliging to connect, and constitutes seamless, high-resolution a image, is had
The resolution ratio higher than single image and the bigger visual field.
At present, in the prior art, the general splicing using joining method and distinguished point based based on region
Method carries out Panorama Mosaic.
Wherein, it is the gray value based on image to be spliced based on region joining method, in image to be matched one
The region of the same size in block region and reference picture calculates ash using least square method or other mathematical methods
The difference of angle value, it is final to realize splicing by comparing, judging the similarity of image overlapping region to be spliced.
Image can also be transformed to by frequency domain by time domain by FFT, then be matched again.
The joining method of feature based causes characteristics of image by image pixel, and standard is characterized as with this, to figure
As the character pair region of lap scans for matching.The method mainly has two processes：Feature extraction
And characteristic matching.The method is estimated using the image split-joint method of feature based using the obvious characteristic of image
Conversion between meter image, it is not necessary to using the full detail of image, the feature of this kind of obvious characteristic such as image
Point (angle point or key point), profile and some not bending moments etc..
Inventor has found the equal Shortcomings part of existing both approaches in process of the present invention is implemented：
Joining method based on region can be attributed to the feature point set for solving image, in image to be spliced
Some characteristic points are chosen, these characteristic points are directed at, that is, realizes the registration of two width figures.However, these features
Point often will choose initial matching point by artificial, which decrease the speed of algorithm, it is impossible to adapt to big data
The image co-registration of amount.The joining method of feature based, it is using feature block matching algorithm, i.e., accurate using correlation
Image registration method then, because correlation method is a kind of searching method of global optimizing, amount of calculation is quite big, spells
Connect speed slower, it is difficult to ensure the real-time of algorithm.
To sum up, it is known that above two joining method has splicing speed slowly, the not strong shortcoming of real-time.
The content of the invention
The technical problems to be solved by the invention are, there is provided a kind of vehicle-mounted panoramic image seamless splicing side
Method and system, splice mapping table, it is possible to achieve many captured by multiple cameras by providing a target image
Width image it is seamless spliced, while have image mosaic speed fast, real-time, and overlapping region will not deposit
The characteristics of obvious distortion.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of vehicle-mounted panoramic image seamless splicing
Method, methods described comprises the following steps：
Primal environment image is obtained by the camera for being arranged at automobile surrounding respectively；
Mapping table is spliced by a target image for previously generating, inquiry obtains each target in target image
Map information corresponding to pixel, wherein, the map information at least including primal environment image sequence number,
The pixel coordinate information of primal environment image；
According to the map information of each target pixel points in the target image, the original of corresponding sequence number is obtained
Pixel value in ambient image on respective pixel coordinate, and obtain each on target image according to the pixel value
The final pixel value of target pixel points.
Wherein, the step of previously generating target image splicing mapping table is further included, it is specifically included：
Target image is divided into multiple target areas according to world coordinate system, and will be positioned at two target areas
Region in the preset range of intersection is defined as splicing regions, and determines each target area and splice region
Primal environment picture numbers corresponding to domain；
According to the mapping relations between primal environment image coordinate system and world coordinate system, and target image
Mapping relations between coordinate system and world coordinate system, obtain on the target image each target pixel points with
The map information of original image maps mutually, wherein, in the splicing regions of target image, each object pixel
A specific pixel point maps mutually on point primal environment image different from two width respectively；
By the positional information of each target pixel points of the target image and each target pixel points are corresponding reflects
The information of penetrating is preserved, and obtains target image splicing mapping table.
Wherein, according to the map information of each target pixel points in the target image, corresponding sequence number is obtained
Pixel value on primal environment image on respective pixel coordinate, and obtained on target image according to the pixel value
The step of final pixel value of each target pixel points, further include：
For the target pixel points of target area, the primal environment figure that each target pixel points is mapped is obtained
As the pixel value on upper respective pixel coordinate, and as the final pixel value of the target pixel points；
For the target pixel points of splicing regions, two original rings that its each pixel is mapped are obtained respectively
Pixel value in the figure of border in corresponding pixel points, and according to predetermined weight to right in the two width primal environment figure
Answer pixel value on pixel to be calculated, obtain the final pixel value of the target pixel points.
Wherein, the pixel value in corresponding pixel points in the two width primal environment figure is entered according to predetermined weight
The step of row calculating, final pixel value for obtaining the target pixel points, is specially：
The final pixel value of target pixel points is obtained by following formula：
Wherein, d is the sequence number of target pixel points, and s1 and s2 is corresponding pixel points in two width primal environment figures
Sequence number；F1 and f2 are predefined weight, between two borders of itself and the pixel and the splicing regions away from
From associated.
Wherein, further include：
Brightness to splicing regions image is adjusted.
Correspondingly, the embodiment of the present invention also provides a kind of vehicle-mounted panoramic image seamless splicing system, is used for
By several primal environment image mosaics into a width panorama target image, it is characterised in that the system includes：
Primal environment image acquisition unit, for obtaining original respectively by being arranged at the camera of automobile surrounding
Ambient image；
Query unit, the target image for being previously generated by splices mapping table, and inquiry obtains target figure
Map information as in corresponding to each target pixel points, wherein, the map information at least includes original
The sequence number of ambient image, the pixel coordinate information of primal environment image；
Splicing unit, for the map information according to each target pixel points in the target image,
The pixel value on respective pixel coordinate on the primal environment image of corresponding sequence number is obtained, and according to the pixel value
Obtain the final pixel value of each target pixel points on target image.
Wherein, further include that target image splices mapping table generation unit, specifically include：
Target image division unit, for target image to be divided into multiple target areas according to world coordinate system,
And the region being located in two preset ranges of target area intersection is defined as splicing regions, and determine every
Primal environment picture numbers corresponding to one target area and splicing regions；
Mapping relations determining unit, for according to reflecting between primal environment image coordinate system and world coordinate system
The mapping relations penetrated between relation, and the coordinate system and world coordinate system of target image, obtain the target
The map information of each target pixel points and original image maps mutually on image, wherein, in target image
In splicing regions, a specific picture on each target pixel points primal environment image different from two width respectively
Vegetarian refreshments maps mutually；
Mapping table generation unit, for by the positional information of each target pixel points of the target image and often
The corresponding map information of one target pixel points is preserved, and obtains target image splicing mapping table.
Wherein, the splicing unit is further included：
Target area processing unit, for the target pixel points for target area, obtains each target picture
Pixel value on the primal environment image that vegetarian refreshments is mapped on respective pixel coordinate, and as the object pixel
The final pixel value of point；
Splicing regions processing unit, for the target pixel points for splicing regions, obtains its each picture respectively
Pixel value in the two width primal environment figures that vegetarian refreshments is mapped in corresponding pixel points, and according to predetermined weight pair
Pixel value in the two width primal environment figure in corresponding pixel points is calculated, and obtains the target pixel points
Final pixel value.
Wherein, the splicing regions processing unit is further included：
Computing unit, the final pixel value for obtaining target pixel points by following formula：
Wherein, d is the sequence number of target pixel points, and s1 and s2 is corresponding pixel points in two width primal environment figures
Sequence number；F1 and f2 are predefined weight, between two borders of itself and the pixel and the splicing regions away from
From associated.
Wherein, further include：
Splice region brightness adjustment unit, is adjusted for the brightness to splicing regions image.
Implement the present invention, with following beneficial effect：
First, the vehicle-mounted panoramic image seamless splicing processing method that the present invention is provided, it is possible to achieve multiple shooting
Multiple image captured by head is spliced, and forms panorama birds-eye view, its visual effect and actual top view
Effect is basically identical；
Meanwhile, the present invention provide embodiment, by the coordinate according to each pixel in target image with
The coordinate information of correspondence primal environment image carries out map information, forms generation target image splicing mapping table；
Carry out panoramic picture it is seamless spliced when, only need to by search the target image splice mapping table, you can to obtain
On the corresponding primal environment image that each target pixel points are mapped on target image on respective pixel coordinate
Pixel value, it is fast, real-time with image mosaic speed such that it is able to be rapidly completed the seamless spliced of panoramic picture
The strong advantage of property, reduces requirement and complexity of the seamless spliced processing system of panoramic picture to hardware；；
In addition, the region in the target image mutually splicing two primal environment images, is set to splicing regions；
In the splicing regions, the pixel of its image is derived from two adjacent cameras, i.e., counted respectively by weight
The pixel value of its corresponding pixel in two neighboring camera original image is calculated, target image is formed most
Whole pixel value, meanwhile, brightness adjustment can be carried out to the image of splicing regions；Such that it is able to improve splice region
The validity and display effect of area image, it is to avoid the phenomenon for image fault and deformity occur occurs.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to implementing
Example or the accompanying drawing to be used needed for description of the prior art are briefly described, it should be apparent that, describe below
In accompanying drawing be only some embodiments of the present invention, for those of ordinary skill in the art, do not paying
On the premise of going out creative labor, other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 is a kind of master of vehicle-mounted panoramic image seamless splicing processing method one embodiment that the present invention is provided
Schematic flow sheet；
Fig. 2 be the present invention provide a kind of vehicle-mounted panoramic image seamless splicing processing method in generate target image
Splice the schematic flow sheet of mapping table；
Fig. 3 is the schematic diagram of the one embodiment for carrying out region division in Fig. 2 to target image；
Fig. 4 is the schematic diagram of another embodiment for carrying out region division in Fig. 2 to target image；
Fig. 5 is a kind of knot of vehicle-mounted panoramic image seamless splicing system one embodiment that the present invention is provided
Structure schematic diagram；
Fig. 6 is the structural representation of target image splicing mapping table generation unit in Fig. 5；
Fig. 7 is the structural representation of splicing unit in Fig. 5.
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 clearly
Chu, it is fully described by, it is clear that described embodiment is only a part of embodiment of the invention, rather than
Whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art are not making creation
Property work under the premise of the every other embodiment that is obtained, belong to the scope of protection of the invention.
As shown in figure 1, being a kind of one reality of vehicle-mounted panoramic image seamless splicing processing method that the present invention is provided
Apply the main flow schematic diagram of example.In this embodiment, the method can just several primal environment image mosaics
Into the target image of a width panorama, specifically, the method includes the steps：
Step S10, primal environment image (i.e. camera is obtained by the camera for being arranged at automobile surrounding respectively
What is obtained looks down picture)；Specifically, four width original is obtained by being arranged at automobile four cameras all around
Beginning ambient image, each width primal environment image to that should have a sequence number, each width primal environment image
The information such as coordinate position, the pixel value of pixel can be recorded；
Step S12, mapping table is spliced by a target image for previously generating, every in inquiry acquisition target image
Map information corresponding to one target pixel points, the map information at least sequence including primal environment image
Number, the information such as the pixel coordinate of primal environment image；Wherein, target image is needs what is spliced and show
Panoramic picture (final spliced aerial view), is stored with target image in target image splicing mapping table
Corresponding relation between sequence number, the pixel coordinate of each pixel and primal environment image；
Step S14, according to the map information of each target pixel points in the target image, obtains corresponding sequence
Number primal environment image on pixel value on respective pixel coordinate, and target figure is obtained according to the pixel value
As the final pixel value of upper each target pixel points, and the final pixel value is assigned to corresponding target picture
Vegetarian refreshments, and show, so that image content of the display corresponding to primal environment image in the target image；Wherein,
Target image splicing mapping table can record the map information of each target pixel points, i.e., can record each target picture
The sequence number of the corresponding primal environment image of vegetarian refreshments, and the pixel coordinate on the primal environment image of the sequence number,
Therefore splice mapping table by inquiring about target image, you can it is corresponding to each target pixel points maps mutually to obtain
Respective pixel coordinate on the primal environment image of sequence number, so as to obtain the pixel value of the respective pixel coordinate；
It is understood that in some other embodiment, may further include step S16 (in Fig. 1 not
Show), the brightness to splicing regions image is adjusted, seamless spliced complete so as to be generated on target image
Scape view.
As shown in Fig. 2 in showing a kind of vehicle-mounted panoramic image seamless splicing processing method of present invention offer
Previously generate the schematic flow sheet that target image splices mapping table；Wherein, generation target image splicing mapping table
Process specifically include the steps：
Step S20, multiple target areas are divided into by target image according to world coordinate system, and will be positioned at two
Region in the preset range of target area intersection is defined as splicing regions, and determine each target area with
And the sequence number of the primal environment image corresponding to splicing regions；
For ease of understanding, refer to shown in Fig. 3, it illustrates carrying out one of region division to target image
The schematic diagram of embodiment, in this embodiment, in the target image, will be in automobile region all around
Tetra- regions of specific F, B, L, R are marked off, wherein, point P1~P4 is four tops of the surrounding of automobile
Point, and the line segment that is made up of point P3, P4 and form two of target image by the line segment that point P5, P6 are constituted
Border, the specific of h0~h3, w1, H_CAR in target image, W_CAR etc. can be obtained by measurement
Range information, wherein, H_CAR is the length of automobile image in target image, and W_CAR is target figure
The width of automobile image as in, it is to be understood that the physical length of length and automobile according to the H_CAR
Deng scaling translation relation, may be used to determine pixel and pixel in primal environment image in target image
Coordinate between mapping relations.Simultaneously, it may be determined which the pixel value of the pixel in each region comes from
Width primal environment image, for example, the pixel value of the pixel in F regions can come from the (figure of camera before automobile
The camera is represented in 3 with the constitutional diagram of small circle and small square frame, similarly hereinafter) captured by primal environment figure
Picture, the pixel value of the pixel in Zone R domain is from the primal environment image captured by car right side camera, L areas
The pixel value of the pixel in domain is from the primal environment image captured by automobile left side camera, the pixel in B regions
The pixel value of point is from the primal environment image captured by automotive back camera；Captured by different cameras
Primal environment image can be made a distinction by different ambient image sequence number i.
As shown in figure 4, the schematic diagram of another embodiment that region division is carried out to target image is shown,
In this embodiment, in the target image, except by automobile region division all around go out specific F,
Tetra- regions of B, L, R, the both sides of splicing seams also between two width primal environment images, according to equal angular
(in figure be a angles) expands certain limit respectively, with formed the A1 regions shown in splicing regions, i.e. figure,
A2 regions, A3 regions and A4 regions.For the pixel of the pixel in these splicing regions in target image
Value can be by calculating its corresponding pixel in the primal environment image captured by two neighboring camera
Pixel value is obtained, and this process is referred to as the fusion process of pixel in the present invention, hereinafter can be detailed
Description.
Step S22, according to the mapping relations between primal environment image coordinate system and world coordinate system, and mesh
Mapping relations between the coordinate system and world coordinate system of logo image, obtain each picture on the target image
The map information of vegetarian refreshments and primal environment image maps mutually, wherein it is possible to understand, in target image
In splicing regions, due to needing to carry out the fusion of pixel, each pixel is different from two width original respectively
A specific pixel point maps mutually in ambient image；For example, in target image the pixel in A1 regions picture
Element value is, it is necessary to a specified point respectively in picture according to captured by automobile front side camera and left side camera
The pixel of pixel carry out calculating acquisition；
Specifically, note primal environment image coordinate is (u0i,v0i), target image coordinate is (u, v), the target figure
As the mapping relations of upper each pixel and original image are primal environment image coordinate (u0i,v0i) and environment
Mapping relations between picture numbers i and target image coordinate (u, v).Take world coordinates (xw,yw,zw) as middle
Amount, finds out (u, v) and (u respectively0i,v0i) the two is with its mapping relations, it is hereby achieved that (u, v) and (u0i,v0i) it
Between mapping relations.
Specifically, in one example, the coordinate mapping relations of primal environment image determine by imaging model,
For example, the relation between camera coordinate system and coordinates of original image coordinates system, can use such as scara models,
Mapping relations are as follows：
Camera coordinate<->Actual imaging coordinate:
Wherein, Xci, Yci, Zci refer to the coordinate of camera, and Ai represents affine transformation matrix；fiIn () represents
The multinomial of outer nonlinearity in parameters projection function Taylor series expansion.
And mapping relations of the target image (top view) and world coordinates between are relatively simple, world coordinates is arrived
Only by scaling, the conversion process of translation between top view coordinate.In short, target image and body of a motor car
It is scaling relation that surrounding needs the region of display, therefore by the coordinate of target image, can calculate corresponding vehicle body
Coordinate, then joined by outside camera, corresponding camera coordinate is calculated, finally according to camera internal reference, meter
Calculate the coordinate of primal environment image.
During above-mentioned Coordinate Conversion, world coordinates (x is mapped to from (u, v)w,yw,zw) after, you can according to
(xw,yw,zw) belonging to region (F, L, R, B) determine the value of the sequence number i of ambient image.
Step S24, by the positional information of each pixel of the target image and each pixel is corresponding reflects
The information of penetrating is preserved, and obtains target image splicing mapping table.
With reference to Fig. 4, some steps in Fig. 1 are described in detail.
Further included in step S14：
For the target pixel points of target area, the primal environment figure that each target pixel points is mapped is obtained
As the pixel value on upper respective pixel coordinate, and as the final pixel value of the target pixel points；
For the target pixel points of splicing regions, the two width original that its each target pixel points is mapped is obtained respectively
Pixel value in beginning environment map in corresponding pixel points, and according to predetermined weight to the two width primal environment figure
Pixel value in middle corresponding pixel points is calculated, and obtains the final pixel value of the target pixel points, the mistake
Journey can be simply referred to as two fusion process of pixel pixel value；
Specifically, the pixel value in corresponding pixel points in two width primal environment figures is calculated by following formula,
Obtain the final pixel value of target pixel points：
Wherein, d is the sequence number of target pixel points, and s1 and s2 is corresponding pixel points in two width primal environment figures
Sequence number；F1 and f2 are the predefined weight of each pixel, and it is two with the pixel and the splicing regions
The distance between border is associated.
Specifically, for example, in the A1 of region, there is a point Pd, itself and the angle border in F regions
Distance is 1, and is 9 with the distance on the angle border in L regions, it is assumed that s1 is corresponding automobile front side
The corresponding pixel points in primal environment image captured by camera；Assuming that s2 is imaged for corresponding automobile left side
The corresponding pixel points in primal environment image captured by head；In one embodiment, can be by can f1 values
It is 9, and is 1 by f2 values, the final pixel value of Pd points can be obtained by above formula.By this
The method of sample, can take corresponding weight according to pixel with the relation on the distance between two angle borders,
Apart from smaller, then weight is bigger, and the image of splicing regions can be enable to be reduced very well, will not go out
The excessive phenomenon of existing distortion.
Meanwhile, in step s 16, i.e., wrap the step of brightness and color to splicing regions image are adjusted
Include：
Precalculate the adjusting parameter of each camera；
The final pixel value of each pixel of the splicing regions is carried out using the adjusting parameter of the camera
Adjustment.
It is understood that because the complexity and stability of color integrated regulation are difficult to hold, it is assumed that different
White balance is consistent between camera image, and its difference is main on overall brightness.Namely RGB three primary colors are believed
For number, it is believed that R, G, B color component ratio are consistent between different cameras, slightly have not only in intensity
Together.So without individually processing tri- passages of R/G/B, single image regulation parameter only has one.
Because the Y-component of yuv data represents brightness of image, therefore only can calculate regulation parameter on Y passages.
In one example, the adjusting parameter of each camera can be calculated using following methods：
To four splicing regions of splicing seams both sides (A1, A2, A3 and A4 region are labeled as in Fig. 4,
In other examples, can with Direct Mark as A, B, C, D region), it is counted respectively and is taken the photograph two neighboring
It is A as the value in head original imageF、AL、BF、BR、CL、CT、DR、DTIf, four tune of camera
Whole parameter is respectively JF、JL、JR、JT, then ε=(J is madeFAF-JLAL)2+ (JFBF-JRBR)2+ (JLCL-JTCT)2+ (JRDR-JTDT)2
It is minimum, you can obtain the adjusting parameter of each camera of needs.
It is understood that in other examples, it is also possible to obtain each camera in other manners
Adjusting parameter, it is for instance possible to use the mean flow rate of adjacent camera obtains each of the splicing regions
The adjusting parameter of camera.
Refer to shown in Fig. 5, a kind of vehicle-mounted panoramic image seamless splicing system of present invention offer is provided
The structural representation of one embodiment of uniting；Refer to Fig. 6 and Fig. 7 in the lump simultaneously, in this embodiment, this is vehicle-mounted
The seamless spliced processing system 1 of panoramic picture can by several primal environment image mosaics into a width panorama target
Image, it includes：
Primal environment image acquisition unit 10, for obtaining original respectively by being arranged at the camera of automobile surrounding
Beginning ambient image；
Query unit 12, the target image for being previously generated by splices mapping table, and inquiry obtains target
Map information in image corresponding to each target pixel points, wherein, the map information at least includes original
The sequence number of beginning ambient image, the pixel coordinate information of primal environment image；
Splicing unit 14, for the map information according to each target pixel points in the target image,
The pixel value on respective pixel coordinate on the primal environment image of corresponding sequence number is obtained, and according to the pixel value
The final pixel value of each target pixel points on target image is obtained, and the final pixel value is assigned to phase
The target pixel points answered, and show.
Target image splicing mapping table generation unit 16, for generating target image splicing mapping table, the mesh
The positional information of each target pixel points of target area and the splicing regions of being stored with logo image splicing mapping table
And the corresponding mapping relations of each pixel and primal environment image；
Splice region brightness adjustment unit 18, is adjusted for the brightness to splicing regions image.
Wherein, target image splicing mapping table generation unit 16 is specifically included：
Target image division unit 160, for target image to be divided into multiple target areas according to world coordinate system
Domain, and the region being located in two preset ranges of target area intersection is defined as splicing regions, and really
Primal environment picture numbers corresponding to fixed each target area and splicing regions；
Mapping relations determining unit 162, for according between primal environment image coordinate system and world coordinate system
Mapping relations between mapping relations, and the coordinate system and world coordinate system of target image, obtain the mesh
The map information of each pixel and original image maps mutually in logo image, wherein, in the spelling of target image
In connecing region, a specific pixel point phase on each pixel primal environment image different from two width respectively
Mapping；
Mapping table generation unit 164, for by the positional information of each target pixel points of the target image and
Each target pixel points are preserved with the corresponding map information of primal environment image, are obtained target image and are spelled
Connect mapping table.
The splicing unit 14 is further included：
Target area processing unit 140, for the target pixel points for target area, obtains each target
Pixel value on the primal environment image that pixel is mapped on respective pixel coordinate, and as the target picture
The final pixel value of vegetarian refreshments；
Splicing regions processing unit 142, for the target pixel points for splicing regions, obtains its each respectively
Pixel value in the two width primal environment figures that target pixel points are mapped in corresponding pixel points, and according to predetermined
Weight is calculated the pixel value in corresponding pixel points in the two width primal environment figure, obtains the target
The final pixel value of pixel.
The splicing regions processing unit 142 is further included：
Computing unit 144, the final pixel value for obtaining target pixel points by following formula：
Wherein, d is the sequence number of target pixel points, and s1 and s2 is corresponding pixel points in two width primal environment figures
Sequence number；F1 and f2 are predefined weight, between two borders of itself and the pixel and the splicing regions away from
From associated.
More details may be referred to the foregoing description to Fig. 1 to Fig. 4, not be described in detail herein.
Implement the present invention, with following beneficial effect：
First, the vehicle-mounted panoramic image seamless splicing processing method that the present invention is provided, it is possible to achieve multiple shooting
Multiple image captured by head is spliced, and forms panorama birds-eye view, its visual effect and actual top view
Effect is basically identical；
Meanwhile, the present invention provide embodiment, by the coordinate according to each pixel in target image with
The coordinate information of correspondence primal environment image carries out map information, forms generation target image splicing mapping table；
Carry out panoramic picture it is seamless spliced when, only need to by search the target image splice mapping table, you can to obtain
On the corresponding primal environment image that each target pixel points are mapped on target image on respective pixel coordinate
Pixel value, it is fast, real-time with image mosaic speed such that it is able to be rapidly completed the seamless spliced of panoramic picture
The strong advantage of property, reduces requirement and complexity of the seamless spliced processing system of panoramic picture to hardware；
In addition, the region in the target image mutually splicing two primal environment images, is set to splicing regions；
In the splicing regions, the pixel of its image is derived from two adjacent cameras, i.e., counted respectively by weight
The pixel value of its corresponding pixel in two neighboring camera original image is calculated, target image is formed most
Whole pixel value, meanwhile, brightness adjustment can be carried out to the image of splicing regions；Such that it is able to improve splice region
The validity and display effect of area image, it is to avoid the phenomenon for image fault and deformity occur occurs.
Above disclosed is only a kind of preferred embodiment of the invention, can not limit this hair with this certainly
Bright interest field, therefore the equivalent variations made according to the claims in the present invention, still belong to what the present invention was covered
Scope.
Claims (10)
1. a kind of vehicle-mounted panoramic image seamless splicing processing method, it is characterised in that methods described includes as follows
Step：
Primal environment image is obtained by the camera for being arranged at automobile surrounding respectively；
Mapping table is spliced by a target image for previously generating, inquiry obtains each target in target image
Map information corresponding to pixel, wherein, the map information at least including primal environment image sequence number,
The pixel coordinate information of primal environment image；
According to the map information of each target pixel points in the target image, the original ring of corresponding sequence number is obtained
Pixel value on the image of border on respective pixel coordinate, and each mesh on target image is obtained according to the pixel value
Mark the final pixel value of pixel.
2. a kind of vehicle-mounted panoramic image seamless splicing processing method as claimed in claim 1, it is characterised in that
The step of previously generating target image splicing mapping table is further included, it is specifically included：
Target image is divided into multiple target areas according to world coordinate system, and will be positioned at two target areas
Region in the preset range of intersection is defined as splicing regions, and determines each target area and splice region
The sequence number of the primal environment image corresponding to domain；
According to the mapping relations between primal environment image coordinate system and world coordinate system, and target image
Mapping relations between coordinate system and world coordinate system, obtain on the target image each target pixel points with
The map information of original image maps mutually, wherein, in the splicing regions of target image, each object pixel
A specific pixel point maps mutually on point primal environment image different from two width respectively；
By the positional information of each target pixel points of the target image and each target pixel points are corresponding reflects
The information of penetrating is preserved, and obtains target image splicing mapping table.
3. a kind of vehicle-mounted panoramic image seamless splicing processing method as claimed in claim 2, it is characterised in that
According to the map information of each target pixel points in the target image, the primal environment figure of corresponding sequence number is obtained
As the pixel value on upper respective pixel coordinate, and each target picture on target image is obtained according to the pixel value
The step of final pixel value of vegetarian refreshments, further include：
For the target pixel points of target area, the primal environment image that each target pixel points are mapped is obtained
Pixel value on upper respective pixel coordinate, and as the final pixel value of the target pixel points；
For the target pixel points of splicing regions, the two width original that its each target pixel points is mapped is obtained respectively
Pixel value in beginning environment map in corresponding pixel points, and according to predetermined weight to the two width primal environment figure
Pixel value in middle corresponding pixel points is calculated, and obtains the final pixel value of the target pixel points.
4. a kind of vehicle-mounted panoramic image seamless splicing processing method as claimed in claim 3, it is characterised in that
It is described the pixel value in corresponding pixel points in the two width primal environment figure is counted according to predetermined weight
The step of calculation, final pixel value for obtaining the target pixel points, is specially：
The final pixel value of target pixel points is obtained by following formula：
Wherein, d is the sequence number of target pixel points, and s1 and s2 is corresponding pixel points in two width primal environment figures
Sequence number；F1 and f2 are predefined weight, between two borders of itself and the pixel and the splicing regions away from
From associated.
5. a kind of vehicle-mounted panoramic image seamless splicing processing method as described in any one of claim 2 to 4,
Characterized in that, further including：
Brightness to splicing regions image is adjusted.
6. a kind of vehicle-mounted panoramic image seamless splicing system, it is characterised in that the system includes：
Primal environment image acquisition unit, for obtaining original respectively by being arranged at the camera of automobile surrounding
Ambient image；
Query unit, the target image for being previously generated by splices mapping table, and inquiry obtains target figure
Map information as in corresponding to each target pixel points, wherein, the map information at least includes original
The sequence number of ambient image, the pixel coordinate information of primal environment image；
Splicing unit, for the map information according to each target pixel points in the target image, obtains
The pixel value on respective pixel coordinate on the primal environment image of corresponding sequence number is obtained, and is obtained according to the pixel value
Obtain the final pixel value of each target pixel points on target image.
7. a kind of vehicle-mounted panoramic image seamless splicing system as claimed in claim 6, it is characterised in that
Further include that target image splices mapping table generation unit, specifically include：
Target image division unit, for target image to be divided into multiple target areas according to world coordinate system,
And the region being located in two preset ranges of target area intersection is defined as splicing regions, and determine every
Primal environment picture numbers corresponding to one target area and splicing regions；
Mapping relations determining unit, for according to reflecting between primal environment image coordinate system and world coordinate system
The mapping relations penetrated between relation, and the coordinate system and world coordinate system of target image, obtain the target
The map information of each target pixel points and original image maps mutually on image, wherein, in the spelling of target image
In connecing region, a specific pixel on each target pixel points primal environment image different from two width respectively
Point maps mutually；
Mapping table generation unit, for by the positional information of each target pixel points of the target image and often
The corresponding map information of one target pixel points is preserved, and obtains target image splicing mapping table.
8. a kind of vehicle-mounted panoramic image seamless splicing system as claimed in claim 7, it is characterised in that
The splicing unit is further included：
Target area processing unit, for the target pixel points for target area, obtains each target picture
Pixel value on the primal environment image that vegetarian refreshments is mapped on respective pixel coordinate, and as the object pixel
The final pixel value of point；
Splicing regions processing unit, for the target pixel points for splicing regions, obtains its each mesh respectively
Pixel value in the two width primal environment figures that mark pixel is mapped in corresponding pixel points, and according to predetermined power
The pixel value in corresponding pixel points in the two width primal environment figure is calculated again, obtains the target picture
The final pixel value of vegetarian refreshments.
9. a kind of vehicle-mounted panoramic image seamless splicing system as claimed in claim 8, it is characterised in that
The splicing regions processing unit is further included：
Computing unit, the final pixel value for obtaining target pixel points by following formula：
Wherein, d is the sequence number of target pixel points, and s1 and s2 is corresponding pixel points in two width primal environment figures
Sequence number；F1 and f2 are predefined weight, between two borders of itself and the pixel and the splicing regions away from
From associated.
10. a kind of vehicle-mounted panoramic image seamless splicing system as described in claim any one of 7-9, its
It is characterised by, further includes：
Splice region brightness adjustment unit, is adjusted for the brightness to splicing regions image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510903646.5A CN106856000B (en) | 2015-12-09 | 2015-12-09 | Seamless splicing processing method and system for vehicle-mounted panoramic image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510903646.5A CN106856000B (en) | 2015-12-09 | 2015-12-09 | Seamless splicing processing method and system for vehicle-mounted panoramic image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106856000A true CN106856000A (en) | 2017-06-16 |
| CN106856000B CN106856000B (en) | 2021-02-05 |
Family
ID=59132295
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510903646.5A Active CN106856000B (en) | 2015-12-09 | 2015-12-09 | Seamless splicing processing method and system for vehicle-mounted panoramic image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106856000B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023080A (en) * | 2016-05-19 | 2016-10-12 | 沈祥明 | Seamless splicing processing system for vehicle-mounted panoramic image |
| CN107633545A (en) * | 2017-08-25 | 2018-01-26 | 包谦 | A kind of processing method of panoramic picture |
| CN109615583A (en) * | 2018-11-30 | 2019-04-12 | 北京龙创悦动网络科技有限公司 | A kind of generation method and device of map |
| CN110341597A (en) * | 2018-04-02 | 2019-10-18 | 杭州海康威视数字技术股份有限公司 | A kind of vehicle-mounted panoramic video display system, method and Vehicle Controller |
| CN110428361A (en) * | 2019-07-25 | 2019-11-08 | 北京麒麟智能科技有限公司 | A kind of multiplex image acquisition method based on artificial intelligence |
| CN110910312A (en) * | 2019-11-21 | 2020-03-24 | 北京百度网讯科技有限公司 | Image processing method and device, automatic driving vehicle and electronic equipment |
| CN110930312A (en) * | 2018-09-19 | 2020-03-27 | 驭势（上海）汽车科技有限公司 | Method and device for generating fisheye camera image |
| CN111242842A (en) * | 2020-01-15 | 2020-06-05 | 深圳市中天安驰有限责任公司 | Image conversion method, terminal and storage medium |
| CN113658058A (en) * | 2021-07-22 | 2021-11-16 | 武汉极目智能技术有限公司 | Brightness balancing method and system in vehicle-mounted all-round system |
| US20220327659A1 (en) * | 2021-04-08 | 2022-10-13 | Raytheon Company | Mitigating transitions in mosaic images |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629372A (en) * | 2012-02-22 | 2012-08-08 | 北京工业大学 | 360 degree panoramic aerial view generation method used for assisting vehicle driving |
| 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 |
| CN103607606A (en) * | 2013-12-02 | 2014-02-26 | 合一网络技术(北京)有限公司 | Video view estimating method and device based on word network |
| CN103763479A (en) * | 2013-12-31 | 2014-04-30 | 深圳英飞拓科技股份有限公司 | Splicing device for real-time high speed high definition panoramic video and method thereof |
-
2015
- 2015-12-09 CN CN201510903646.5A patent/CN106856000B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102629372A (en) * | 2012-02-22 | 2012-08-08 | 北京工业大学 | 360 degree panoramic aerial view generation method used for assisting vehicle driving |
| 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 |
| CN103607606A (en) * | 2013-12-02 | 2014-02-26 | 合一网络技术(北京)有限公司 | Video view estimating method and device based on word network |
| CN103763479A (en) * | 2013-12-31 | 2014-04-30 | 深圳英飞拓科技股份有限公司 | Splicing device for real-time high speed high definition panoramic video and method thereof |
Non-Patent Citations (1)
| Title |
|---|
| 赵凯: "全景可视化辅助泊车***研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106023080A (en) * | 2016-05-19 | 2016-10-12 | 沈祥明 | Seamless splicing processing system for vehicle-mounted panoramic image |
| CN107633545A (en) * | 2017-08-25 | 2018-01-26 | 包谦 | A kind of processing method of panoramic picture |
| CN110341597B (en) * | 2018-04-02 | 2020-11-27 | 杭州海康威视数字技术股份有限公司 | Vehicle-mounted panoramic video display system and method and vehicle-mounted controller |
| CN110341597A (en) * | 2018-04-02 | 2019-10-18 | 杭州海康威视数字技术股份有限公司 | A kind of vehicle-mounted panoramic video display system, method and Vehicle Controller |
| CN110930312A (en) * | 2018-09-19 | 2020-03-27 | 驭势（上海）汽车科技有限公司 | Method and device for generating fisheye camera image |
| CN109615583A (en) * | 2018-11-30 | 2019-04-12 | 北京龙创悦动网络科技有限公司 | A kind of generation method and device of map |
| CN110428361A (en) * | 2019-07-25 | 2019-11-08 | 北京麒麟智能科技有限公司 | A kind of multiplex image acquisition method based on artificial intelligence |
| CN110910312A (en) * | 2019-11-21 | 2020-03-24 | 北京百度网讯科技有限公司 | Image processing method and device, automatic driving vehicle and electronic equipment |
| CN111242842A (en) * | 2020-01-15 | 2020-06-05 | 深圳市中天安驰有限责任公司 | Image conversion method, terminal and storage medium |
| CN111242842B (en) * | 2020-01-15 | 2023-11-10 | 江苏中天安驰科技有限公司 | Image conversion method, terminal and storage medium |
| US20220327659A1 (en) * | 2021-04-08 | 2022-10-13 | Raytheon Company | Mitigating transitions in mosaic images |
| US11769224B2 (en) * | 2021-04-08 | 2023-09-26 | Raytheon Company | Mitigating transitions in mosaic images |
| CN113658058A (en) * | 2021-07-22 | 2021-11-16 | 武汉极目智能技术有限公司 | Brightness balancing method and system in vehicle-mounted all-round system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106856000B (en) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106856000A (en) | A kind of vehicle-mounted panoramic image seamless splicing processing method and system | |
| CN106023080A (en) | Seamless splicing processing system for vehicle-mounted panoramic image | |
| CN101276465B (en) | Method for automatically split-jointing wide-angle image | |
| CN106056536A (en) | Vehicle-mounted panorama image seamless splicing processing method | |
| CN103226822B (en) | Medical imaging joining method | |
| CN104794683B (en) | Based on the video-splicing method scanned around gradual change piece area planar | |
| CN107424120A (en) | A kind of image split-joint method in panoramic looking-around system | |
| CN104732482A (en) | Multi-resolution image stitching method based on control points | |
| US20160005316A1 (en) | Around view system and operating method thereof | |
| CN107154014A (en) | A kind of real-time color and depth Panorama Mosaic method | |
| CN111768332B (en) | Method for splicing vehicle-mounted panoramic real-time 3D panoramic images and image acquisition device | |
| CN109883433B (en) | Vehicle positioning method in structured environment based on 360-degree panoramic view | |
| CN110033411A (en) | The efficient joining method of highway construction scene panoramic picture based on unmanned plane | |
| CN108257089B (en) | A method of the big visual field video panorama splicing based on iteration closest approach | |
| CN103985133A (en) | Search method and system for optimal splicing lines among images based on graph-cut energy optimization | |
| CN104463778A (en) | Panoramagram generation method | |
| CN110322485A (en) | A kind of fast image registration method of isomery polyphaser imaging system | |
| US9734611B2 (en) | Around view system and operating method thereof | |
| CN105976324B (en) | Vehicle image splicing method | |
| CN109146791B (en) | Tunnel spread map generation method based on area array CCD imaging | |
| CN115170810B (en) | Visible light infrared image fusion target detection example segmentation method | |
| CN108447042B (en) | Fusion method and system for urban landscape image data | |
| CN103167247B (en) | A kind of video sequence color image joining method | |
| CN106627373B (en) | A kind of image processing method and system for intelligent parking | |
| CN103310482A (en) | Three-dimensional reconstruction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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 |