Research on image binarization algorithm based on evolutionary algorithms
基于演化算法的图像二值化算法研究*

In order to overcome some shortcomings of traditional binarization methods—long search time,low image fidelity,this paper used evolutionary algorithm,because it had some advantages—short search time,high image sincerity,searching in the global space.Using evolutionary algorithm could easily get the image binarization threshold value.Taking the license plate image for example,according to the image complex characteristics,designed encode,crossover and mutation evolution operators of image evolution chromosome,then established the fitness function of solving image binarization value.The data experiments show that,compared with the traditional algorithm,using evolutionary algorithm for the binarization threshold value can shorten search time,and improve image fidelity,and can do image recognition and image processing more efficiently.