基于人眼视觉的红外图像广义直方图增强算法  

A generalized histogram enhancement algorithm of infrared image based on human visual properties

作　　者：王才子[1] 孙士保[1] 段建辉[1] 完颜丹丹[1] 

机构地区：[1]河南科技大学电子信息工程学院,河南洛阳471003

出　　处：《合肥工业大学学报（自然科学版）》2012年第9期1208-1211,1292,共5页Journal of Hefei University of Technology：Natural Science

基　　金：洛阳市应用技术研究与开发资金资助项目(1101018A)

摘　　要：文章针对传统直方图均衡增强算法存在红外图像细节信息丢失问题,且算法没有考虑人眼视觉特性,提出基于人眼视觉的红外图像广义直方图增强算法;根据像素邻域信息建立广义直方图,进行直方图均衡化处理;在均衡化过程中,利用人眼灰度分辨能力,对拉伸过大的灰度级进行限制,再对被过度压缩的灰度级进行适当拉伸以扩大其动态范围;实验结果表明,该方法有效增强图像局部细节,解决了红外图像视觉模糊的问题,使增强后图像视觉效果更自然。As the traditional histogram equalization enhancement algorithms have the problems of the loss of the infrared image details and the exclusion of humans visual properties, a generalized histogram enhancement algorithm of the infrared image based on human visual properties is proposed. According to the pixel neigh- borhood information, a generalized histogram is established. Then the histogram equalization processing is carried out. During the equalization processing, the gray-scale that is overstretched is limited on the basis of the gray-scale resolution capabilities of the human eyes, and the gray-scale with over-compression is stretched properly to expand its dynamic range. The experimental results show that this method can enhance the local details of the image, solve the problem of the blurred vision of the infrared image and achieve a more natural visual effect of the image after enhancement.

关 键 词：红外图像 图像增强 人眼视觉特性 广义直方图 直方图均衡 

分 类 号：TN911.73[电子电信—通信与信息系统]