基于遗传算法的最佳熵阈值图像分割法

Entropic Thresholding Method Based on Genetic Algorithm

将遗传算法用于图像分割的Ｋａｐｕｒ 等人提出的最佳熵阈值确定法（ 简称ＫＳＷ 熵法） 中，进行了针对图像分割遗传程序所需的参数设计．ＫＳＷ 熵方法具有很多优点，但同时也存在弱点：需要大量的运算时间，特别是在计算多阈值时．因此需要引入优化算法．Ｊ． Ｈｏｌｌａｎｄ 的遗传算法是具有鲁棒性和自适应性的搜索方法．采用遗传算法实现单阈值和多阈值图像分割，实验结果表明分割速度快于传统的ＫＳＷ 熵法，缩短了运算时间．

The method of entropic thresholding proposed by Kapur, Sahoo and Wong (KSW) is implemented using Genetic Algorithm(GA). Optimum parameters suitable for this algorithm are also given. KSW method has many advantages. However, it has weakness: it needs a great deal of computational time especially when computing multithreshold. So it needs to import optimization technique. Genetic Algorithm proposed by John Holland is a robust and adaptive stochastic searching method. Entropic single thresholding and multithresholding methods are all presented using GA. The results show that it can shorten the computational time compared with the classical KSW method.