Entropic Thresholding Method Based on Genetic Algorithm
-
摘要: 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.Abstract: 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.
-
Key words:
- image processing /
- entropy /
- threshold value
-
1. Sahoo P K, Soltani S, Wong A K C .A survey of thresholding techniques. Computer Vision, Graphics, and Image Processing,1988,41:233~260 2. Kapur J N, Sahoo P K, Wong A K C .A new method of gray-level picture thresholding using the entropy of the histogram. Computer Vision, Graphics, and Image Processing,1985, 29:273~285 3. Lance C. Practical handbook of genetic algorithm Volume I: applications.Boca Raton:CRC Press, 1995 4. Zbigniew M. Genetic algorithm + data structure = evolution programs. 3rd ed.Berlin:Springer Verlag, 1996
点击查看大图
计量
- 文章访问数: 3727
- HTML全文浏览量: 246
- PDF下载量: 2654
- 被引次数: 0