Automatic white balance method and device

Abstract

The invention can eliminate color distortion of object caused by that the image contains object with bulk of high saturation, and can adjust yield value of white balance of three base color components: red, green, and blue under each condition of light source. The method includes steps: dividing the captured image into pixel block; determining light emitting source of the image; the step for iterative computing yield value of white balance is in use for obtaining a suitable yield value of white balance; using the said suitable yield value of white balance to carry out adjustment of white balance for pixel points in entire image. The device for automatic white balance includes the image segmentation module, the module for determining light emitting source of image, the module for iterativecomputing yield value of white balance, and the image correction module.

Description

Automatic white balance method and device technical field The invention relates to a fast iterative automatic white balance method based on prior knowledge. The invention also relates to a device for realizing the automatic white balance method. Background technique The human visual system is capable of shifting light sources by ensuring that white objects appear white. Under different light source environments, the intensities of the three primary color components of red, green, and blue vary greatly, and the blue component in sunlight is more than that in cool white fluorescent light. The human visual system balances the components of red, green, and blue light to ensure that white objects appear white in both daylight and fluorescent light. This technique is called white balancing. The automatic white balance device in the image acquisition system uses the automatic white balance algorithm to simulate the human visual system, so as to reproduce the true white of the ...

AI Technical Summary

Problems solved by technology

The disadvantage of using a single white pixel criterion is that it cannot be used to determine the illuminant source of the image, because a single white pixel criterion for all types of illumination sources does not have enough information to support further analysis to obtain an equivalent illuminant source

In addition, the above method does not have a fast response to the change of the light source, because the adjustment process of the color gain value will be very long

Images