首页 | 本学科首页   官方微博 | 高级检索  
     

基于最大熵原则和灰度变换的图像增强
引用本文:章秀华,杨坤涛.基于最大熵原则和灰度变换的图像增强[J].光电工程,2007,34(2):84-87,119.
作者姓名:章秀华  杨坤涛
作者单位:华中科技大学,光电子科学与工程学院,湖北,武汉,430074;华中科技大学,光电子科学与工程学院,湖北,武汉,430074
摘    要:提出了一种利用最大熵原则和灰度变换进行图像对比度增强的方法.在最大熵原则基础上利用条件迭代算法对图像灰度级进行最佳分类,对各分类区域进行相应的灰度变换,根据不同需要选取变换参数,在图像对比度增强同时各区域均衡性也得到很大改善.将利用条件迭代算法计算最大熵多阈值的方法与最小均方误差(LMSE)计算多阈值的方法进行比较,实验结果表明,文中所用方法在迭代次数上大大低于基于最小均方误差算法所需迭代次数,节省了图像处理时间,图像均衡化效果也相对提高.

关 键 词:图像增强  最大熵原则  灰度变换  区域均衡
文章编号:1003-501X(2007)02-0084-04
收稿时间:2006/2/27
修稿时间:2006-02-272006-05-26

Image enhancement based on maximum entropy principle and gray-level transformation
ZHANG Xiu-hua,YANG Kun-tao.Image enhancement based on maximum entropy principle and gray-level transformation[J].Opto-Electronic Engineering,2007,34(2):84-87,119.
Authors:ZHANG Xiu-hua  YANG Kun-tao
Affiliation:Department of Optoelectronic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China
Abstract:An image contrast enhancement method based on maximum entropy principle and gray-level transformation was proposed. According to the maximum entropy principle, the image gray levels were classified optimally by iterated conditional modes, and the transformation parameters were selected to enhance the image contrast and improve region homogenization. The method of iterated conditional modes was compared with Least Mean Square Error (LMSE) method in computing multi-thresholds. The experiment result demonstrates that compared with LMSE, the proposed method spends less time and acquires greater image region homogenization result.
Keywords:Image enhancement  Maximum entropy principle  Gray-level transformation  Region homogenization
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号