CN101588439B - Method for correcting multi-anaglyph color cast - Google Patents

Abstract

The invention relates to a method for color cast correction of multiple parallax maps, which is characterized in that an image with true color is selected as a standard map in multiple parallax maps, and then the standard parallax map and the color cast parallax map are respectively converted from RGB to YUV color space conversion, through the three color correction coefficients K _Y , K _U , K _V of all pixels in the YUV color space through the color cast disparity map, the real color gray value of the color cast image to be corrected in the YUV color space is obtained, and finally Convert the obtained real color gray value from YUV to RGB color space, so as to realize the correction of any number of color cast parallax images. The invention can realize the color cast correction of arbitrary multi-path parallax images before synthesizing stereoscopic images, can improve the effect of autostereoscopic displays, and the method is simple and feasible.

Description

A Color Cast Correction Method for Multiple Parallax Maps

1. Technical field

The invention relates to stereo shooting technology and digital image processing technology, especially for color correction of multi-parallax images captured by array cameras.

2. Technical background

Three-dimensional stereoscopic display has always been the dream of human beings. Among them, the multi-view autostereoscopic display based on binocular parallax is an important three-dimensional stereoscopic display. The viewer feels the real three-dimensional effect. Its display content comes from scene images captured from multiple angles, which can be generated by 3D modeling software, or multiple disparity maps captured by camera arrays from different angles. In order to increase the stereoscopic viewing area, the number of cameras that need to be installed is usually relatively large, generally as many as 8, or even more than a dozen. However, camera differences are inevitable, even if they are from the same brand, such as CCD color filters, camera lens aperture settings, etc., cannot be completely consistent, which will affect the color of the disparity map obtained by shooting. At the same time, because the camera is placed in different positions when shooting, the shooting environment will also have different effects on each lens. Affected by all the above factors, there will be color differences in the images captured by each camera in the camera array, and sometimes the color differences will be large. When multiple disparity maps are synthesized, it will affect the quality of the composite image. , which not only affects the three-dimensional effect of the display, but also causes visual fatigue to the viewer in severe cases. Therefore, in order to give full play to the performance of the multi-view autostereoscopic display, it is very necessary to perform color correction on the disparity map before synthesizing the stereoscopic images.

3. Contents of the invention

Therefore, the purpose of the present invention is to solve the problem of color difference between the multi-disparity images captured by the camera array, so that the stereoscopic display has a more perfect stereoscopic display effect. The present invention selects one of the standard parallax maps without color cast as a reference among multiple disparity maps, and sequentially performs color correction on other color cast parallax maps.

First, convert the standard disparity map and the color cast disparity map from the RGB color space to the YUV color space, and the conversion formula from the RGB color space to the YUV color space is expressed as:

y=0.299×r+0.587×g+0.114×b

u＝(-0.169)×r+(-0.332)×g+0.500×b+128 (1)

v=0.500×r+(-0.419)×g+(-0.0813)×b+128

In formula (1), r, g, and b respectively represent the gray value of a certain pixel in the RGB color space in a disparity map, and y, u, and v respectively represent the gray value of the pixel in the YUV color space after conversion value.

Secondly, the correction coefficients K _Y , K U , and K _V of the Y, _U , and V channels are calculated respectively.

It is assumed that the size of the standard disparity map and the color cast disparity map is m×n. The sum of the Y-channel grayscale values of all pixels in the standard disparity map is denoted as Y ₁ , and the Y-channel grayscale value of any pixel is denoted as y _1mn . The sum of the values of the Y channels of all pixels in the color cast disparity map is denoted as Y ₂ , and the value of the Y channel of any pixel is denoted as y _2mn . but:

${\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}\mathrm{<span\; class="notranslate">\; mn</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 2</span>}<span\; class="notranslate">\; )</span>$

${\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; the\; y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; mn</span>}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 3</span>}<span\; class="notranslate">\; )</span>$

Suppose the ratio of Y ₁ to Y ₂ is denoted as K _Y , and the subscript y is expressed as the Y channel correction coefficient of the color cast parallax map, then:

${\mathrm{<span\; class="notranslate">\; K</span>}}_{\mathrm{<span\; class="notranslate">\; Y</span>}}<span\; class="notranslate">\; =</span>\frac{{\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; 1</span>}}}{{\mathrm{<span\; class="notranslate">\; Y</span>}}_{\mathrm{<span\; class="notranslate">\; 2</span>}}}<span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; 4</span>}<span\; class="notranslate">\; )</span>$

Multiply the correction coefficient K _Y of the Y channel with the value y _2mn of each pixel Y channel in the color cast disparity map to obtain the corrected Y channel value y′ _2mn of each pixel, namely:

y′ _2mn =K _Y ×y _2mn (5)

In the formula (5), y _2mn represents the value of the Y channel of a certain pixel in the color cast disparity map before correction, and y′ _2mn represents the value of the Y channel of the pixel after correction.

Referring to the same process as above, the correction coefficients of the U and V channels of the corresponding color cast parallax map are K _U and K _V respectively, so that the U and V channel correction of the color cast parallax map can be realized.

Then, after the color correction of the color cast disparity map is completed, the color space conversion from YUV to RGB is performed on the corrected disparity map. The conversion formula is as follows:

r=y+1.4075×(v-128)

g＝y+(-0.3445)×(u-128)+(-0.7169)×(v-128) (6)

b=y+1.7790×(u-128)

In formula (6), y, u, and v respectively represent the gray value of a certain pixel in the corrected disparity map in the YUV color space, and r, g, and b represent the gray value of the pixel in the RGB color space after color space conversion . The conversion from the YUV color space of the corrected disparity map to the RGB color space is completed, and the color correction of a color cast disparity map is completed.

By repeating the above process, color correction for all other color cast disparity maps can be realized.

4. Description of drawings

FIG. 1 is a schematic flowchart of a color cast correction method for multiple parallax images according to the present invention.

5. Specific implementation

The present invention is further described below, specifically related to an 8-camera array multi-camera system, all cameras are arranged in a straight line, the resolution of each camera is m×n=640×480, the color is color, and the bit depth is 8 bits , the shooting speed is the video speed. It is necessary to point out that this embodiment is only used for further description of the present invention, and cannot be interpreted as limiting the protection scope of the present invention, and those skilled in the art make some non-essential improvements to the present invention according to the above-mentioned content of the present invention And adjustments still belong to the protection scope of the present invention. The specific implementation process of a multi-parallax map color cast correction method of the present invention is:

1. Use the 8-camera system to shoot any scene, and then select a disparity map with true color from the 8 disparity maps obtained from the shooting as the standard disparity map;

2. Choose a disparity map with color cast from the remaining 7 disparity maps;

3. Convert the standard disparity map and the color cast disparity map from the RGB color space to the YUV color space using formula (1) to obtain their y, u, and v values;

4. Use formulas (2) and (3) to sum the values of the three channels of Y, U, and V in each pixel of the standard parallax map and the color cast parallax map;

5. Use the formula (4) to obtain the ratio of the total values of the standard parallax map and the three channels of the color cast parallax map Y, U, and V respectively, and obtain the color correction coefficients K _Y , K _U , and K _V ;

6. Use the obtained Y, U, V three-channel correction coefficients K _Y , K _U , K _V to multiply the values of Y, U, and V of each pixel of the color cast parallax map respectively to obtain the YUV color space Corrected color cast disparity map;

7. Convert the corrected color cast disparity map from YUV color space to RGB color space by using the formula (6), and obtain the r, g, b values of the corrected image.

8. Repeat the above process to complete the color correction of other disparity maps in sequence.

So far, a color cast correction method for multi-disparity images has been introduced. In this way, multi-viewpoint parallax images with color cast eliminated can be obtained, which can be used for multi-viewpoint autostereoscopic display.

Claims (1)

1. A method for color cast correction of multiple parallax images, which adopts computer digital image processing technology to carry out color cast correction to the multiple parallax images obtained by the stereo camera system, and is characterized in that a color is selected in multiple parallax images to be true The image is used as a standard image to correct the color cast of other disparity maps, and then the standard disparity map and the color cast disparity map are converted from RGB to YUV color space, and then the standard disparity map and the color cast disparity map are respectively calculated in YUV color The sum of the YUV channel values of all pixels in the space, and calculate the ratio of the sum of their YUV channel values, respectively, to obtain the color correction coefficients K _Y , K _U , K _V of the three channels, and then obtain Y, U The three correction coefficients K _Y , K _U , and K V of the three channels of , _V and the values of the Y, U, and V channels of each pixel of the color cast disparity map are respectively multiplied to obtain the color cast image to be corrected in the YUV color space The real color gray value, finally, convert the obtained real color gray value from YUV to RGB color space, so as to realize the correction of any number of color cast disparity maps.

Images