Automatic white balance method and system for color cast of a large number of pure-color scenes, and terminal

Abstract

The invention provides an automatic white balance method and system for color cast of a large number of pure-color scenes, and a terminal. The method comprises the steps: comparing the RGB values of small blocks of an image through an image segmentation method, determining whether an original image is a pure-color background or not, and further carrying out the region division of the original image, judging whether the shot original image is the pure-color background or not more accurately, and for the original image with the pure-color background, obtaining the most accurate white balance gain through weighted average of the automatic white balance gain calculated by the small image blocks and the automatic white balance gain calculated by the whole original image.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an automatic white balance method, system and terminal for color cast of a large number of pure color scenes. Background technique [0002] In the traditional white balance method, in color correction, the original image must have a gray point or a white point in order to calibrate the original image. A simple color means only one color, such as pure red, then there will be no white point in the original image. , there is no white point as a reference, so the existing white balance algorithm cannot calibrate the pure color image. [0003] Some existing algorithms, such as advanced white balance algorithm: take gray card images at different color temperatures (D75, D65, D50, CWF, H, A) in the light box, fill the gray card with the whole live image, and then calibrate each one. The white balance reference point at the color temperature, and then determine the ...

AI Technical Summary

Problems solved by technology

[0004] Another example is the grayscale world algorithm: count all pixels in the live scene as white points, count Ravg, Bavg, and Gavg of all pixels, and then use Gavg / Ravg and Gavg / Bavg as gain compensation for R and B channels respectively ; This algorithm does not consider the recognition of a single solid-color scene, and does not consider the situation where there are no gray points or few gray points in the shot scene. When a large-area solid-color scene has almost no gray points or few gray points, it will render the non-gray The point is misjudged as a gray point, and the white balance correction is performed statistically

[0005] Therefore, the above two algorithms have a high probability of color cast in pure color scenes

Images