Background Color Constancy Algorithm Based on Neural Network

摘要

This paper proposed a novel color constancy algorithm based on neural network to solve the problems of color constancy in the field of machine vision such as background image update. Because of the generalization capability of neural network, by means of appropriate sample sets, we adopt improved learning algorithm to train the neural network to obtain the mapping relation of the corresponding pixels of the image before and after the changes. After training, the neural network would output image data with color constancy. At last, the method was tested by the experiment of the video background update under the scenes with randomly selected illuminants, and it was proved to be effective.