文档介绍：第 34 卷第 3 期光电工程 ,
2007 年 3 月 Opto-Electronic Engineering March, 2007
文章编号:1003-501X(2007)03-0068-04
模拟频域滤波重构直方图均衡的图像增强方法
刘焕雨 1,2,万秋华 1,熊文卓 1,楚广生 1,吴永芝 1,2
( 1. 中国科学院长春光学精密机械与物理研究所,吉林长春 130033;
2. 中国科学院研究生院,北京 100039 )
摘要:针对传统图像增强方法的不足,提出一种基于模拟频域滤波重构直方图均衡的图像增强方法。将频率域滤
波的思想引入空间域直方图运算当中,在模拟频率直方图统计中进行频率信息的统计,利用这些频率信息建立模
拟频率域坐标,进行模拟频域滤波,对滤波处理后的直方图进行均衡化处理。实验表明:与传统方法相比,该算
法优化了灰度级的动态分布范围,能得到更清晰的增强效果,且在图像中没有视觉明显的噪声放大。
关键词:图像增强;模拟频率直方图统计;模拟频域滤波;重构直方图均衡
中图分类号: 文献标识码:A
Image enhancing method based on reconstructed histogram equalization
after simulated frequency domain filter
LIU Huan-yu1, 2,WAN Qiu-hua1,XIONG Wen-zhuo1,CHU Guang-sheng1,WU Yong-zhi1, 2
( 1. Changchun Institute of Optics, Fine Mechanics and Physics,
the Chinese Academy of Sciences, Changchun 130033, China;
2. Graduate School of the Chinese Academy of Sciences, Beijing 100039, China )
Abstract:In view of the faults of traditional image enhancing method, a new method based on reconstructed histogram
equalization after simulating frequency domain filter is presented. Frequency domain filter algorithm is applied to spatial
domain histogram operation. Frequency information is acquired statistically in simulated frequency histogram statistic
first, then simulated frequency domain coordinates are constituted with this frequency information, and simulated
frequency domain filter is done. Last, histogram filtered is equalized. Experimental results show that, compared with
general methods, the new way can optimize dynamic range of the gray level, enhance image more clearly, and amplify
noise unobviously in processed image.
Key words:Image enhancement; Simulated frequency histogram statistic; Simulated frequency filter; Reconstructed
histogram equalization
引言
图像增强技术在医学、工业检测、地貌学以及军事等多个领域都有着极其广泛的应用。由于成像系统
成像器件的灵敏度、分辨率和噪声特性以及环境照度等因素,致使图像存在模糊、噪声大等缺点。为改善
图像的视觉质量,通常需要对图像进行对比度增强处理。改善图像的质量应不丢失图像细节,不引入过大
噪声及不引起细节失真。此外,为提高方法的实用性,