一种图像增强的混合算法研究  

A Hybrid Algorithm for Image Enhancement

作　　者：张龙涛[1] 孙玉秋[1] 

机构地区：[1]长江大学信息与数学学院,湖北荆州434023

出　　处：《长江大学学报（自科版）（上旬）》2015年第9期30-33,4,共4页JOURNAL OF YANGTZE UNIVERSITY (NATURAL SCIENCE EDITION) SCI ＆ ENG

基　　金：湖北省自然科学基金重点项目(2013CFA053)

摘　　要：针对直方图均衡化方法增强图像时往往会同时增强噪声的问题,提出了一种新混合增强算法:首先对图像进行直方图均衡;然后对经直方图均衡处理后的图像做梯度锐化;最后将锐化后的图像进行灰度扩展,增大图像对比度。选取X射线成像和卫星图像作为待处理图像,分别用2种不同的方法以及笔者提出的方法对其进行处理,然后将处理后的图像进行对比。试验结果表明,该混合算法比其他几种方法效果更好,不仅增加了动态范围,增强了对比度,还丰富了图像细节,而且不会增强噪声。同时还分别用4邻域对比度、8邻域对比度、平均灰度和方差这4个参数对原图和处理后的图像进行对比。从对比结果可见,该算法可以显著提高图像对比度;对于整体灰度较低的图像,该算法还可以提高图像整体灰度;从方差对比结果来看,经该算法处理后,图像质量得到显著提高。Aiming at the problem that noise is raised when the images are enhanced by histogram equalization,a new hybrid algorithm is proposed.Firstly,the images are processed by histogram equalization algorithm.Then,the processed images are enhanced by gradient sharpening method.Finally,the gray extend algorithm is used to process the images to increase their contrast.X-ray images and satellite images are used as the images for processing.These images are enhanced by other two algorithms and the proposed algorithm at the same time.The processed images are compared with each other.The results show that the proposed algorithm in the paper is better than the other algorithms.It can enhance the dynamic range,increase the contrast and present the detail of the images.Furthermore,the proposed algorithm can restrain noise effectively.Also the parameters of four methods,such as 4-neighberhood contrast,8-neighberhood contrast,average gray and variance,are used to compare the processed images.It is shown from the results that the algorithm proposed can increase the contrast of the images and enhance the gray images efficiently,especially for low gray images.

关 键 词：直方图均衡 梯度锐化 灰度扩展 图像处理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]