CN110443771A - It is vehicle-mounted to look around panoramic view brightness and colour consistency method of adjustment in camera system - Google Patents
It is vehicle-mounted to look around panoramic view brightness and colour consistency method of adjustment in camera system Download PDFInfo
- Publication number
- CN110443771A CN110443771A CN201910757941.2A CN201910757941A CN110443771A CN 110443771 A CN110443771 A CN 110443771A CN 201910757941 A CN201910757941 A CN 201910757941A CN 110443771 A CN110443771 A CN 110443771A
- Authority
- CN
- China
- Prior art keywords
- channel
- eye view
- birds
- adjustment
- camera
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 69
- 230000000007 visual effect Effects 0.000 claims abstract description 8
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 239000003086 colorant Substances 0.000 claims description 5
- 235000004522 Pentaglottis sempervirens Nutrition 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000006978 adaptation Effects 0.000 abstract description 4
- 230000008878 coupling Effects 0.000 abstract description 2
- 238000010168 coupling process Methods 0.000 abstract description 2
- 238000005859 coupling reaction Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 16
- 238000005286 illumination Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 240000004050 Pentaglottis sempervirens Species 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Color Television Image Signal Generators (AREA)
- Image Processing (AREA)
Abstract
The present invention, which provides, a kind of vehicle-mounted looks around panoramic view brightness and colour consistency method of adjustment in camera system, comprising the following steps: acquires image by looking around camera system；Image is transformed into YCbCr color space；It determines the zone of mutual visibility domain between image and removes exterior point；To minimize zone of mutual visibility domain luminosity error as the color of target adjustment image and brightness；Generate panoramic view.The brightness of panoramic view different piece and color can be carried out global adaptation by this method, desalinate the splicing seams of panoramic view, improve the visual effect for looking around image, has robustness high, the low feature of cost is calculated, can satisfy the requirement of the real-time in practical application to vehicle-mounted viewing system splicing panoramic view.Furthermore, method proposed by the present invention, with the hardware device for looking around camera system use and look around that the degree of coupling of connection scheme is low, can adapt to it is different look around camera system software and hardware scheme, and be widely used in the looking around in camera system of various types of vehicles or other equipment carrying.
Description
Technical field
The invention belongs to intelligent driving field, it is related to vehicle-mounted looking around camera system technology.
Background technique
With the development of automobile industry, the security performance of vehicle is increasingly taken seriously, therefore many DAS (Driver Assistant System)s
It is all equipped with and vehicle-mounted looks around camera system.Vehicle-mounted camera system of looking around is by being arranged in 4-6 wide-angle camera of vehicle periphery
Be spliced into vehicle periphery looks around image, and the information of road surface of vehicle periphery can be provided to driver, and auxiliary driver goes on a journey
Judgement is sailed, security performance of the vehicle in complex environment is improved.However, the light environment as locating for different cameral in system with
And the depth of exposure of camera itself is different, it is existing vehicle-mounted to look around the image of looking around that camera system is spliced into and be usually present difference
There is larger difference in the brightness in region and color.The brightness of panoramic view different zones and color difference not only will affect ring
The visual effect of view but will adversely affect the judgement of driver's road pavement situation.Therefore, camera is looked around to vehicle-mounted
It is very necessary that the various pieces for the panoramic view that system generates, which carry out brightness and colour consistency adjustment,.
Currently, the method for commonly looking around brightness of image and colour consistency adjustment is broadly divided into two classes: the first kind is pair
Adjacent two images find suitable splicing seams, and merge to the part near splicing seams；Second class is to looking around camera system
Every image does global adaptation respectively in system.The representative method of first kind method has: " Y.C.Liu, K.Y.Lin, Y.S.Chen,
“Bird’s-Eye View Vision System for Vehicle Surrounding Monitoring,”in RobVis
2008. ", this method can find an optimal stitching line when splicing panoramic view to each frame image, each to panoramic view later
Part is merged, but visual continuity of this method in dynamic image is bad, and calculation amount is larger；" Yang Li, Zhu Zhu, Liu
A kind of Ji embedded type automobile of woods " gets a bird's eye view Panoramic Image Mosaic Algorithm, " journal of Zhejiang university (engineering version) 2 (2014): 292-296. "
To region of the adjacent image near splicing seams, weight is distributed according to the distance between pixel and splicing seams, is weighted and averaged
Fusion, the disadvantages of the method are as follows algorithm complexity is higher, and there are larger differences for the whole visual effect of panoramic view different zones
It is different.The representative method of second class method has: " Y.Liu, B.Zhang. " Photometric alignment for surround
View camera system, " in Proc.IEEE ICIP 2014:1827-1831. " is to the tones of panoramic view various pieces
Mapping curve is parameterized, and reduces the luminosity difference between adjacent image, the party by solving optimal hue adjustment parameter
Method effect is more satisfactory, but iteration optimization is needed to solve, and it is larger to calculate cost, it is difficult to reach real-time；"S.R.Han,J.Min,
T.Park,et al.“Photometric and geometric rectification for stereoscopic
Images, " in Proc.SPIE-Int.Soc.Opt.Eng., 2012. " are mended by the global brightness histogram of matching image
Repay the luminance difference of adjacent image, but this method adjacent cameras public domain it is less look around it is ineffective on image, and
And the color of image is not adjusted accordingly.Although the above method can desalinate the splicing looked around in image to a certain extent
Seam improves the quality for looking around image, but existing method there is adjustment effect and calculates the contradiction between cost: effect is good
Method it is often more complicated, it is difficult to meet the requirement of real-time；It is not ideal enough to calculate the low method effect of cost.In other words,
The technique study difficult point for looking around brightness of image and colour consistency adjustment at present is to design one kind and takes into account adjustment effect and in real time
The algorithm of property.
Summary of the invention
It vehicle-mounted look around panoramic view brightness and colour consistency in camera system the purpose of the present invention is to provide a kind of and adjusts
The brightness of panoramic view different piece and color can be carried out global adaptation, desalinate the spelling between different piece by method, this method
Seam improves and looks around the visual effect of image, and it is low to calculate cost, can satisfy in practical application and spells to vehicle-mounted viewing system
Connect the requirement of the real-time of panoramic view.
In order to achieve the above objectives, solution of the invention is:
One kind is vehicle-mounted to look around panoramic view brightness and colour consistency method of adjustment in camera system, comprising the following steps:
(1) image is acquired using each camera looked around in camera system and generate ground birds-eye view respectively；
(2) the ground birds-eye view generated in step (1) is transformed into the color space that luminance channel is separated with Color Channel；
(3) it determines the zone of mutual visibility domain for the ground birds-eye view that adjacent cameras generates in step (2) and removes exterior point；
(4) target is minimised as with the image pixel intensities mean value error in the zone of mutual visibility domain of the adjacent cameras determined in step (3),
The regulation coefficient in each channel in the ground birds-eye view for each camera for asking step (2) to generate；
(5) the ground birds-eye view of each camera generated according to the regulation coefficient generated in step (4), set-up procedure (2)
Each channel pixel intensity value；
(6) ground birds-eye view adjusted in step (5) is converted back into primary colors color space, and according to looking around in camera system
The drawing generating method of looking around used generates panoramic view.
Wherein, the detailed process of each step and main points are as follows:
In step (1):
The camera system of looking around is made of n camera, is respectively { C according to sequence counter-clockwise1, C2..., Cn}；Also,
There are the public visuals field between adjacent cameras；
Any birds-eye view generating algorithm can be used to generate for the ground birds-eye view, respectively { I1, I2..., In, it is preferred that
These birds-eye views should be at same pixel coordinate system OGOn；
The ground birds-eye view should be color RGB image.
In step (2):
The color space space that the luminance channel is separated with Color Channel is YCbCr color space, and wherein the channel Y is bright
Component is spent, the channel Cb is chroma blue component, and the channel Cr is red chrominance component.
Conversion formula of the rgb color space to YCbCr color space are as follows:
The image for being transformed into YCbCr color space is expressed asWherein the channel Y isThe channel Cb isThe channel Cr is
In step (3):
The determination method in the zone of mutual visibility domain for the birds-eye view that adjacent cameras generates is that the ground birds-eye view that adjacent cameras generates exists
Pixel coordinate system OGThe identical part of middle coordinate is zone of mutual visibility domain.For a pair of of adjacent cameras CiAnd Cj, wherein j=(i+1) %
N, % are complementation symbol, and zone of mutual visibility domain representation is
The determination method of exterior point is, for a pair of of adjacent cameras CiAnd Cj, for the zone of mutual visibility determined in step (4-1)
DomainIn any pixel p, if its luminance component in the ground birds-eye view of the corresponding YCbCr color space of two cameras
Absolute value of the difference be greater than specific threshold threshold, i.e.,Point p is determined as exterior point；
After removing in the zone of mutual visibility domain of adjacent image, the zone of mutual visibility domain of rest part is denoted as determining exterior point
In step (4), for each channel C h ∈ { Y, Cb, Cr }, each camera is asked correspondingly using identical method
The regulation coefficient of face birds-eye view, the steps include:
(a) to any pair of adjacent cameras CiAnd Cj, wherein j=(i+1) %n, seeks the ground birds-eye view of their generations respectively
In the zone of mutual visibility domain of removal exterior pointIn image pixel intensities mean value, be denoted as respectivelyWithIts calculation formula is:
WhereinFor the function of statistical pixel number.
(b) camera C is setiThe regulation coefficient of the channel C h of corresponding ground birds-eye view isThen with adjacent after adjustment
The zone of mutual visibility domain image pixel intensities mean value error of the corresponding ground birds-eye view of camera is minimised as target, and can establish an equation group:
(c) bound term is added to the equation group in step (b)It obtains finally solving
Equation group:
Equation group is solved using least square method, the tune of the channel C h of the corresponding ground birds-eye view of each camera can be obtained
Integral coefficient
In step (5), for each channel C h ∈ { Y, Cb, Cr }, getting a bird's eye view for each camera is adjusted using identical method
The pixel intensity value in each channel of figure, method particularly includes:
According to the regulation coefficient of the channel C h of the corresponding ground birds-eye view of the obtained each camera of step (4)Adjust the pixel intensity value of the channel C h of the corresponding ground birds-eye view of each camera.For Ci pairs of camera
The pixel intensity value of the channel C h for the ground birds-eye view answeredIn a point p, method of adjustment are as follows:
WhereinFor the channel C h of ground birds-eye view adjusted.Preferably, three channels Ch ∈ { Y, Cb, Cr } are closed
Ground birds-eye view after and is expressed as
In step (6), for camera CiCorresponding ground birds-eye view adjustedConvert it back to original color sky
Between, i.e. its main points of rgb color space are as follows:
Conversion formula of the YCbCr color space to rgb color space are as follows:
The image of rgb color space after conversion is expressed asWherein the channel R isThe channel G isChannel B is
For being transformed into the image of primary colors color spaceIn any point p, for its any channel H ∈ { R, G, B },
IfThen willIt is assigned a value of 255；IfThen willIt is assigned a value of 0.
In step (6), it is described look around drawing generating method and can be used arbitrarily look around drawing generating method, specifically: to look around
The image of each camera acquisition can generate corresponding ground birds-eye view, and adjacent phase as input respectively in camera system
There should be zone of mutual visibility domain between the corresponding birds-eye view of machine.
By adopting the above-described technical solution, the invention has the following advantages:
The present invention, which adjusts the YCbCr color space that brightness of image and color use, has the characteristics that brightness and color separated,
Agree with goal in research of the invention, and it is linear transformation relationship between rgb color space, mutually conversion calculates cost
Low, speed is fast.
The method that the present invention has used global adaptation using adjustment brightness of image and color, in the premise for guaranteeing adjustment effect
Under, a parameter is only distributed to each channel for looking around each part in image, advantage is low to calculate cost, and robustness
It is high.
The method for the rejecting exterior point that the present invention is used when solving the zone of mutual visibility domain of adjacent image, can efficiently reduce altogether
Influence of the chaff interferent to integrated regulation result in viewed area, improves the robustness of algorithm.
Panoramic view brightness proposed by the present invention and colour consistency method of adjustment are filled with the hardware for looking around camera system use
The degree of coupling for setting and looking around connection scheme is low, can adapt to it is different look around camera system software and hardware scheme, therefore can be wide
It is looked around in camera system applied to what various types of vehicles or other equipment were carried generally.
Detailed description of the invention
Fig. 1 is the algorithm flow chart of present invention adjustment panoramic view brightness and colour consistency.
Fig. 2 is the schematic diagram in adjacent cameras zone of mutual visibility domain in panoramic view of the present invention.
Fig. 3-A is the Contrast on effect of algorithm proposed by the present invention under the conditions of uniform illumination, and before left side is adjustment, right side is to adjust
After whole.
Fig. 3-B is the Contrast on effect of algorithm proposed by the present invention under intense light conditions, and before left side is adjustment, right side is adjustment
Afterwards.
Fig. 3-C is the Contrast on effect of algorithm proposed by the present invention under strong light intensity shadowed condition, and before left side is adjustment, right side is
After adjustment.
Fig. 3-D be ground exist it is strong it is reflective under the conditions of algorithm proposed by the present invention Contrast on effect, before left side is adjustment,
Right side is after adjusting.
Specific embodiment
Below in conjunction with the embodiment shown in that figure, the present invention will be further described.
The vehicle-mounted specific side for looking around panoramic view brightness and colour consistency method of adjustment in camera system proposed by the present invention
Case is:
One, it generates and looks around image
It includes 4 cameras that camera system is looked around used in the present embodiment, is respectively arranged in all around four of vehicle
Orientation, is the fisheye camera that field range is 185 °, and original resolution is 1920 pixels × 1080 pixels, the image pane of acquisition
Formula is color RGB image, codomain 0-255；It is to start with forward sight camera, is respectively { C according to sequence counter-clockwise1, C2, C3,
C4, and there are the public visuals field between adjacent cameras.What the viewing system generated looks around image " initial panoramic view " as shown in figure 1
Shown, resolution ratio is 1000 pixels × 1000 pixels, pixel coordinate system OG, the upper left corner is coordinate origin, and X-axis is level side
To the right, Y-axis is that vertical direction is downward.
Two, convert color spaces
The image of looking around acquired in step 1 is belonged into C1, C2, C3, C4The birds-eye view separation of four cameras, the bird after separation
Figure of looking down from a height is respectively { I1, I2, I3, I4, as shown in figure 1 shown in " birds-eye view before adjusting ".Wherein I1In OGIn corresponding range are as follows: X-axis
- 1000 pixel of 0 pixel of direction, -250 pixel of 0 pixel of Y direction；I2In OGIn corresponding range are as follows: 0 pixel -250 of X-direction
Pixel, -1000 pixel of 0 pixel of Y direction；I3In OGIn corresponding range are as follows: -1000 pixel of 0 pixel of X-direction, Y direction
- 1000 pixel of 750 pixel；I4In OGIn corresponding range are as follows: -1000 pixel of 750 pixel of X-direction, 0 pixel of Y direction -
1000 pixels.
Then the corresponding birds-eye view of four cameras is transformed into YCbCr color space from rgb color space.Rgb color is empty
Between arrive YCbCr color space conversion formula are as follows:
The image for being transformed into YCbCr color space is expressed asWherein the channel Y isThe channel Cb isThe channel Cr is
Three, it extracts zone of mutual visibility domain and removes exterior point
I in step 2 is determined first1, I2, I3, I4Zone of mutual visibility domain range.I1And I2Zone of mutual visibility domain in OGIn it is corresponding
Range are as follows: -250 pixel of 0 pixel of X-direction, -250 pixel of 0 pixel of Y direction are expressed asI2And I3Zone of mutual visibility domain exist
In OGIn corresponding range are as follows: -250 pixel of 0 pixel of X-direction, -1000 pixel of 750 pixel of Y direction are expressed asI3
And I4Zone of mutual visibility domain in OGIn corresponding range are as follows: -1000 pixel of 750 pixel of X-direction, 750 pixel -1000 of Y direction
Pixel is expressed asI4And I1Zone of mutual visibility domain in OGIn corresponding range are as follows: -1000 pixel of 750 pixel of X-direction, Y
- 250 pixel of 0 pixel of axis direction, is expressed asFig. 2 illustrates the signal in the zone of mutual visibility domain of each camera birds-eye view in the present embodiment
Scheme, " each camera zone of mutual visibility domain " partially illustrates the schematic diagram in the adjacent cameras zone of mutual visibility domain extracted in Fig. 1.
Then exterior point is removed to each zone of mutual visibility domain, the determination method of exterior point is, for a pair of of adjacent cameras CiAnd Cj, wherein 1
≤ i≤4, j=(i+1) %4, determining zone of mutual visibility domainIn any pixel p, if it is in the corresponding YCbCr color of two cameras
The absolute value of the difference of luminance component in the ground birds-eye view of color space is greater than 64, i.e., Point p, that is, true
It is set to exterior point.
After removing in the zone of mutual visibility domain of adjacent image, the zone of mutual visibility domain of rest part is denoted as determining exterior point
Four, regulation coefficient is solved
Image of the present embodiment for the YCbCr color space generated in step 2
Each channel C h ∈ { Y, Cb, Cr }, the regulation coefficient of the corresponding ground birds-eye view of each camera is sought using identical method,
Step are as follows:
(a) to any pair of adjacent cameras CiAnd Cj, wherein 1≤i≤4, j=(i+1) %4, the ground for asking them to generate respectively
Face birds-eye view is in the zone of mutual visibility domain of removal exterior pointIn image pixel intensities mean value, be denoted as respectivelyWithIts calculation formula
Are as follows:
WhereinFor the function of statistical pixel number.
(b) camera I is set1, I2, I3, I4The regulation coefficient of the channel C h of corresponding ground birds-eye view is respectivelyThen with the zone of mutual visibility domain image pixel intensities mean value of the corresponding ground birds-eye view of adjacent cameras after adjustment
It minimizes the error as target, can establish an equation group:
(c) bound term is added to the equation group in step (b)Finally required
The equation group of solution:
Equation group is solved using least square method, the tune of the channel C h of the corresponding ground birds-eye view of each camera can be obtained
Integral coefficient
Five, original image is adjusted
Image of the present embodiment for the YCbCr color space generated in step 2
Each channel C h ∈ { Y, Cb, Cr }, the image pixel intensities in each channel of the birds-eye view of each camera are adjusted using identical method
Value.
According to the regulation coefficient of the channel C h of the corresponding ground birds-eye view of the obtained each camera of step 5Adjust the pixel intensity value of the channel C h of the corresponding ground birds-eye view of each camera.For 1≤i≤
4, camera CiThe pixel intensity value of the channel C h of corresponding ground birds-eye viewIn a point p, method of adjustment are as follows:
WhereinFor the channel C h of ground birds-eye view adjusted.Then, three channels Ch ∈ { Y, Cb, Cr } are merged,
Ground birds-eye view after merging is expressed asAs shown in figure 1 shown in " each camera birds-eye view after adjustment ".
Six, it restores color space and generates panoramic view
The present embodiment is for camera C1, C2, C3, C4Corresponding ground birds-eye view adjustedConvert it back to former rgb color space.YCbCr color space is to rgb color space
Conversion formula are as follows:
The image of rgb color space after conversion is expressed asWherein the channel R is
The channel G isChannel B is
For being transformed into the image of primary colors color spaceIn any point p, it is any logical for its wherein 1≤i≤4
Road H ∈ { R, G, B }, ifThen willIt is assigned a value of 255；IfThen willIt is assigned a value of 0.
Finally, generating panoramic view adjusted according to image connecting method is looked around described in the present embodiment, " adjust as shown in figure 1
Shown in panoramic view after whole ".
Beneficial effects of the present invention are illustrated below by way of specific experiment result shown in Fig. 3:
Fig. 3-A is the Contrast on effect of algorithm proposed by the present invention under the conditions of uniform illumination, and before left side is adjustment, right side is to adjust
After whole.Image is looked around it can be seen that the image global illumination that single camera is shot under the conditions of uniform illumination is uniform, but before adjustment
Between can have overall brightness difference, it is adjusted to look around the luminance difference in image and eliminate.
Fig. 3-B is the Contrast on effect of algorithm proposed by the present invention under intense light conditions, and before left side is adjustment, right side is adjustment
Afterwards.It can be seen that fade effect is presented in the brightness under intense light conditions of forward sight and left view camera, and visible obvious between adjacent cameras
Splicing seams and color difference, image adjusted of looking around remain the former light gradient phenomenon looked around in image, and it is basic
Eliminate the splicing seams and color difference between adjacent image.
Fig. 3-C is the Contrast on effect of algorithm proposed by the present invention under strong light intensity shadowed condition, and before left side is adjustment, right side is
After adjustment.The left view of visible shadow covering, it is right depending on and the brightness of backsight image be significantly higher than the forward sight image without obvious shade, and
Present color difference, it is adjusted after image of looking around each section brightness and color are adjusted to consistent level.
Fig. 3-D be ground exist it is strong it is reflective under the conditions of algorithm proposed by the present invention Contrast on effect, before left side is adjustment,
Right side is after adjusting.It can be seen that there is obvious reflective right view and backsight the image exposure as caused by reflective compensation, brightness is bright
It is aobvious to be lower than left view and forward sight image, it is adjusted after look around image for each section brightness adjustment to consistent level.
Above example contains the scene for leading to panoramic view each section luminance difference common in practical application.Experiment card
Bright, method proposed by the present invention has good effect under various scenes.
The above-mentioned description to examples of implementation is that this can be understood and applied for the ease of those skilled in the art
Invention.Person skilled in the art obviously easily can make various modifications to these embodiments, and illustrating herein
General Principle be applied in other embodiments without having to go through creative labor.Therefore, the present invention is not limited to realities here
Example, those skilled in the art's announcement according to the present invention are applied, improvement and modification made without departing from the scope of the present invention all should
Within protection scope of the present invention.
Claims (12)
- Vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system 1. a kind of, which is characterized in that including with Lower step:(1) image is acquired using each camera looked around in camera system and generate ground birds-eye view respectively；(2) the ground birds-eye view generated in step (1) is transformed into the color space that luminance channel is separated with Color Channel；(3) it determines the zone of mutual visibility domain for the ground birds-eye view that adjacent cameras generates in step (2) and removes exterior point；(4) target is minimised as with the image pixel intensities mean value error in the zone of mutual visibility domain of the adjacent cameras determined in step (3), seeks step Suddenly the regulation coefficient in each channel in the ground birds-eye view for each camera that (2) generate；(5) the ground birds-eye view of each camera generated according to the regulation coefficient generated in step (4), set-up procedure (2) is each The pixel intensity value in channel；(6) ground birds-eye view adjusted in step (5) is converted back into primary colors color space, and is used according to looking around in camera system Look around drawing generating method generate panoramic view.
- 2. it is according to claim 1 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (1):Camera system is looked around described in (1-1) to be made of n camera, is respectively { C according to sequence counter-clockwise1, C2..., Cn}；Any birds-eye view generating algorithm can be used to generate for (1-2) described ground birds-eye view, respectively { I1, I2..., In}；(1-3) described ground birds-eye view should be color RGB image.
- 3. it is according to claim 2 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (1-1): there are the public visuals field between adjacent cameras；Or/andIn step (1-2), birds-eye view is in same pixel coordinate system OGOn.
- 4. it is according to claim 1 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (2):The color space that (2-1) described luminance channel is separated with Color Channel is YCbCr color space, and wherein the channel Y is brightness Component, the channel Cb are chroma blue component, and the channel Cr is red chrominance component；Conversion formula of the rgb color space described in (2-2) to YCbCr color space are as follows:The image that (2-3) is transformed into YCbCr color space is expressed asWherein the channel Y isThe channel Cb isThe channel Cr is
- 5. it is according to claim 1 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, Be characterized in that, step (3) the following steps are included:The determination method in the zone of mutual visibility domain for the birds-eye view that (3-1) adjacent cameras generates is the ground birds-eye view that adjacent cameras generates In pixel coordinate system OGThe identical part of middle coordinate is zone of mutual visibility domain；For a pair of of adjacent cameras CiAnd Cj, wherein j=(i+ 1) %n, % are complementation symbol, and zone of mutual visibility domain representation isThe determination method of (3-2) exterior point is, for a pair of of adjacent cameras CiAnd Cj, for the zone of mutual visibility determined in step (3-1) DomainIn any pixel p, if its luminance component in the ground birds-eye view of the corresponding YCbCr color space of two cameras Absolute value of the difference be greater than specific threshold threshold, i.e.,Point p is determined as exterior point；(3-3) by step (3-2) determine exterior point from step (3-1) determine zone of mutual visibility domain in remove after, rest part Zone of mutual visibility domain be denoted as
- 6. it is according to claim 1 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (4), for each channel C h ∈ { Y, Cb, Cr }, each camera is asked correspondingly using identical method The regulation coefficient of face birds-eye view, the steps include:(4-1) is to any pair of adjacent cameras CiAnd Cj, wherein j=(i+1) %n, the ground birds-eye view for asking them to generate respectively exist Remove the zone of mutual visibility domain of exterior pointIn image pixel intensities mean value, be denoted as respectivelyWithIts calculation formula is:WhereinFor the function of statistical pixel number；(4-2) sets camera CiThe regulation coefficient of the channel C h of corresponding ground birds-eye view isThen with the adjacent phase after adjustment The zone of mutual visibility domain image pixel intensities mean value error of the corresponding ground birds-eye view of machine is minimised as target, and can establish an equation group:Bound term is added to the equation group in step (4-2) in (4-3)It obtains finally solving Equation group:Equation group is solved using least square method, the adjustment system of the channel C h of the corresponding ground birds-eye view of each camera can be obtained Number
- 7. it is according to claim 6 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (5), for each channel C h ∈ { Y, Cb, Cr }, adjusts getting a bird's eye view for each camera using identical method The pixel intensity value in each channel of figure, method are as follows:According to the regulation coefficient of the channel C h of the corresponding ground birds-eye view of obtained each cameraAdjustment The pixel intensity value of the channel C h of the corresponding ground birds-eye view of each camera.
- 8. it is according to claim 7 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, for camera CiThe pixel intensity value of the channel C h of corresponding ground birds-eye viewIn a point p, adjustment side Method are as follows:WhereinFor the channel C h of ground birds-eye view adjusted.
- 9. it is according to claim 8 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, the ground birds-eye view after three channels Ch ∈ { Y, Cb, Cr } are merged is expressed as
- 10. it is according to claim 1 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (6), for camera CiCorresponding ground birds-eye view adjustedConvert it back to original color sky Between, i.e., rgb color space the step of are as follows:Conversion formula of the YCbCr color space described in (6-1) to rgb color space are as follows:(6-2) is for being transformed into the image of primary colors color space described in step (6-1)In any point p, for it One channel H ∈ { R, G, B }, ifThen willIt is assigned a value of 255；IfThen willIt is assigned a value of 0.
- 11. it is according to claim 10 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, the image of the rgb color space after conversion is expressed as in step (6-1)Its The middle channel R isThe channel G isChannel B is
- 12. it is according to claim 1 it is vehicle-mounted look around panoramic view brightness and colour consistency method of adjustment in camera system, It is characterized in that, in step (6), the drawing generating method of looking around is to look around the image of each camera acquisition in camera system as defeated Enter, corresponding ground birds-eye view can be generated respectively, and there should be zone of mutual visibility domain between the corresponding birds-eye view of adjacent cameras.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910757941.2A CN110443771B (en) | 2019-08-16 | 2019-08-16 | Method for adjusting consistency of brightness and color of annular view in vehicle-mounted annular view camera system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910757941.2A CN110443771B (en) | 2019-08-16 | 2019-08-16 | Method for adjusting consistency of brightness and color of annular view in vehicle-mounted annular view camera system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110443771A true CN110443771A (en) | 2019-11-12 |
| CN110443771B CN110443771B (en) | 2023-07-21 |
Family
ID=68436048
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910757941.2A Active CN110443771B (en) | 2019-08-16 | 2019-08-16 | Method for adjusting consistency of brightness and color of annular view in vehicle-mounted annular view camera system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110443771B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461989A (en) * | 2020-04-02 | 2020-07-28 | 深圳普捷利科技有限公司 | Vehicle-mounted image pixel adjusting method, device, equipment and readable storage medium |
| CN112017222A (en) * | 2020-09-08 | 2020-12-01 | 北京正安维视科技股份有限公司 | Video panorama stitching and three-dimensional fusion method and device |
| CN112801885A (en) * | 2020-12-17 | 2021-05-14 | 中国人民解放军63623部队 | Dimming automatic compensation method |
| CN113421183A (en) * | 2021-05-31 | 2021-09-21 | 中汽数据(天津)有限公司 | Method, device and equipment for generating vehicle panoramic view and storage medium |
| CN113689368A (en) * | 2020-05-18 | 2021-11-23 | 上海赫千电子科技有限公司 | Automatic illumination consistency adjusting method applied to vehicle-mounted all-around image |
| CN114430463A (en) * | 2022-04-01 | 2022-05-03 | 青岛大学附属医院 | Optimization method and system for self-adaptive medical-grade image effect |
| CN114663277A (en) * | 2020-12-23 | 2022-06-24 | 比亚迪股份有限公司 | Image processing method, storage medium and intelligent terminal |
| CN117177076A (en) * | 2022-05-24 | 2023-12-05 | 格兰菲智能科技有限公司 | Channel numerical value calculation method, and method, device, equipment and medium for generating look-around graph |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030234866A1 (en) * | 2002-06-21 | 2003-12-25 | Ross Cutler | System and method for camera color calibration and image stitching |
| CN105488777A (en) * | 2015-04-22 | 2016-04-13 | 兰雨晴 | System and method for generating panoramic picture in real time based on moving foreground |
| CN107016647A (en) * | 2017-04-26 | 2017-08-04 | 武汉大学 | Panoramic picture color tone consistency correcting method and system |
| CN109493273A (en) * | 2018-10-09 | 2019-03-19 | 江苏裕兰信息科技有限公司 | A kind of color consistency adjusting method |
| CN109948398A (en) * | 2017-12-20 | 2019-06-28 | 深圳开阳电子股份有限公司 | The image processing method and panorama parking apparatus that panorama is parked |
-
2019
- 2019-08-16 CN CN201910757941.2A patent/CN110443771B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030234866A1 (en) * | 2002-06-21 | 2003-12-25 | Ross Cutler | System and method for camera color calibration and image stitching |
| CN105488777A (en) * | 2015-04-22 | 2016-04-13 | 兰雨晴 | System and method for generating panoramic picture in real time based on moving foreground |
| CN107016647A (en) * | 2017-04-26 | 2017-08-04 | 武汉大学 | Panoramic picture color tone consistency correcting method and system |
| CN109948398A (en) * | 2017-12-20 | 2019-06-28 | 深圳开阳电子股份有限公司 | The image processing method and panorama parking apparatus that panorama is parked |
| CN109493273A (en) * | 2018-10-09 | 2019-03-19 | 江苏裕兰信息科技有限公司 | A kind of color consistency adjusting method |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111461989A (en) * | 2020-04-02 | 2020-07-28 | 深圳普捷利科技有限公司 | Vehicle-mounted image pixel adjusting method, device, equipment and readable storage medium |
| CN113689368A (en) * | 2020-05-18 | 2021-11-23 | 上海赫千电子科技有限公司 | Automatic illumination consistency adjusting method applied to vehicle-mounted all-around image |
| CN113689368B (en) * | 2020-05-18 | 2024-07-19 | 上海赫千电子科技有限公司 | Automatic illumination consistency adjustment method applied to vehicle-mounted looking-around image |
| CN112017222A (en) * | 2020-09-08 | 2020-12-01 | 北京正安维视科技股份有限公司 | Video panorama stitching and three-dimensional fusion method and device |
| CN112801885A (en) * | 2020-12-17 | 2021-05-14 | 中国人民解放军63623部队 | Dimming automatic compensation method |
| CN114663277A (en) * | 2020-12-23 | 2022-06-24 | 比亚迪股份有限公司 | Image processing method, storage medium and intelligent terminal |
| CN113421183A (en) * | 2021-05-31 | 2021-09-21 | 中汽数据(天津)有限公司 | Method, device and equipment for generating vehicle panoramic view and storage medium |
| CN113421183B (en) * | 2021-05-31 | 2022-09-20 | 中汽数据(天津)有限公司 | Method, device and equipment for generating vehicle panoramic view and storage medium |
| CN114430463A (en) * | 2022-04-01 | 2022-05-03 | 青岛大学附属医院 | Optimization method and system for self-adaptive medical-grade image effect |
| CN114430463B (en) * | 2022-04-01 | 2022-06-28 | 青岛大学附属医院 | Optimization method and system for self-adaptive medical-grade image effect |
| CN117177076A (en) * | 2022-05-24 | 2023-12-05 | 格兰菲智能科技有限公司 | Channel numerical value calculation method, and method, device, equipment and medium for generating look-around graph |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110443771B (en) | 2023-07-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110443771A (en) | It is vehicle-mounted to look around panoramic view brightness and colour consistency method of adjustment in camera system | |
| CN105761227B (en) | Underwater picture Enhancement Method based on dark channel prior and white balance | |
| WO2018076732A1 (en) | Method and apparatus for merging infrared image and visible light image | |
| EP2798831B1 (en) | Spectral image processing | |
| US8755597B1 (en) | Smart fusion of visible and infrared image data | |
| CN106485681B (en) | Underwater color image restoration method based on color correction and red channel prior | |
| CN102263985B (en) | Quality evaluation method, device and system of stereographic projection device | |
| EP2355037A1 (en) | Method of and system for determining an average colour value for pixels | |
| CN107895357B (en) | A kind of real-time water surface thick fog scene image Enhancement Method based on FPGA | |
| CN107580163A (en) | A kind of twin-lens black light camera | |
| CN111507930B (en) | Image fusion method, device, storage medium and computer equipment | |
| JP5178361B2 (en) | Driving assistance device | |
| CN105578063A (en) | Image processing method and terminal | |
| CN108364263A (en) | A kind of vehicle-mounted image processing method of SD input high definition output | |
| CN111553862A (en) | Sea-sky background image defogging and binocular stereo vision positioning method | |
| CN107085964B (en) | Vehicular automatic driving system based on image enhancement | |
| CN110298812A (en) | A kind of method and device of image co-registration processing | |
| CN111860632B (en) | Multipath image consistency fusion method | |
| CN109377468A (en) | The pseudo-colours fusion method of infra-red radiation and polarization image based on multiple features | |
| CN110290313B (en) | Method for guiding automatic focusing equipment to be out of focus | |
| CN112435161A (en) | Panoramic all-around image splicing method and system, electronic equipment and storage medium | |
| CN111667446A (en) | Image processing method | |
| CN107644403A (en) | The non-uniform color calibration method of severe environmental conditions hypograph | |
| CN105118032B (en) | A kind of wide method for dynamically processing of view-based access control model system | |
| CN115484712A (en) | Control method and device for tunnel entrance lighting system and storage medium thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |