Automatic white balance method and device

Abstract

The invention provides an automatic white balance method and device. The method comprises the following steps: acquiring N standard environment images with different color temperatures, and acquiring tricolor information of the standard environment images; fitting a black body locus curve by using the N different color temperatures and the tricolor information corresponding to the color temperatures; selecting M calibration points on the black body locus curve, wherein M is greater than or equal to N; specific to the calibration points, acquiring sub-white areas of quadrangles of areas in which the calibration points are positioned; acquiring images to be detected, and determining that statistics points of the images to be detected are white points if the statistics points are positioned in the quadrangles of any sub-white area; and otherwise, determining that the statistics points are non-white points. According to the technical scheme of the invention, white area parameters can be generated automatically, and different white area parameters are generated specific to different front-end equipment, so that a user does not need to configure the white area parameters on the front-end equipment; the waste of labor force is reduced; and errors occurring in a configuring process are avoided.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an automatic white balance method and device. Background technique [0002] The visible light of the human eye is composed of the spectral superposition of 7 colors. When people use the naked eye to view the world, they have basically the same perception of the same color under different light. For example, when the sun rises in the morning, when you look at a white object, you feel that it is white; when you look at a white object under the dim light at night, you feel that it is still white. This is because the human brain has adapted to the color reproduction of objects under different light during the growth of human beings from birth. [0003] However, the front-end equipment of the video surveillance system (such as analog cameras, network cameras, digital cameras, etc.) does not have the adaptability of human eyes. Reddish-yellow, so for the same object, the col...

AI Technical Summary

Problems solved by technology

However, factors such as the photosensitive performance of the image sensor of the front-end equipment and the filter in the optical transmission will affect the white area parameters, that is, different color temperatures will correspond to different white area parameters. Therefore, different front-end equipment needs to be manually configured. The configuration process is very complicated, and it is easy to configure mistakes

Images