计算机与数字工程2019，Vol.47Issue(7)：1805-1809,5.DOI:10.3969/j.issn.1672-9722.2019.07.048
基于最大熵和遗传算法的图像分割方法研究
Study of Image Segmentation Method Based on Maximum Entropy and Genetic Algorithm
余荣泉 1段先华1
作者信息
- 1. 江苏科技大学计算机学院 镇江 212003
- 折叠
摘要
Abstract
Image threshold segmentation is one of the important fields of digital image processing. It is also the key and first step of image analysis and recognition. Compared with other threshold methods,maximum entropy threshold method can make full use of spatial information of pixels and distribution information of image pixel gray. Meanwhile,it also has some disadvantages:high complexity of the algorithm,long operation time when calculates multiple threshold and vague segmentation of the target which con?tains detailed information. To increase the efficiency and accuracy of image segmentation,it combines the maximum entropy segmen?tation method and the genetic algorithm together. Firstly,it discusses the threshold segmentation methods of one-dimensional maxi?mum entropy and two-dimensional maximum entropy respectively. Secondly,it searches the optimal threshold under the genetic al?gorithm process of select,cross and mutation. Lastly,it segments the target image with the optimal threshold. Result of the experi?ment shows that this method is of great segmentation efficiency and the accuracy of the segmentation is greatly improved compared with the traditional maximum entropy segmentation method.关键词
图像分割/遗传算法/阈值/最大熵Key words
image segmentation/genetic algorithm/threshold/maximum entropy分类
信息技术与安全科学引用本文复制引用
余荣泉,段先华..基于最大熵和遗传算法的图像分割方法研究[J].计算机与数字工程,2019,47(7):1805-1809,5.
