基于直方图均衡及融合的红外图像增强技术  

IR Image Enhancement Based on Joint Histogram Equalization with Image Fusion

作　　者：张峰[1] 刘上乾[1] 汪大宝[1] 

机构地区：[1]西安电子科技大学技术物理学院,西安710071

出　　处：《半导体光电》2009年第4期632-635,共4页Semiconductor Optoelectronics

基　　金：国家自然科学基金项目(60377034)

摘　　要：针对红外图像的特点,提出了一种基于直方图变换及图像融合的红外图像增强的新方法。该方法利用平稳小波分解进行噪声抑制,通过自适应直方图均衡提高图像对比度,并利用非线性滤波算法增加新的高频分量,最后进行图像融合以实现图像的增强。该算法克服了传统直方图均衡增强背景而削弱目标以及平台直方图均衡中难以选择平台值的缺点,并大幅度提高了图像的细节。实验结果表明,该算法对红外图像具有很好的增强效果,可以很好地抑制背景,提高目标的对比度,同时使目标的细节更加明显。Through analyzing the features of infrared （IR） images, a new adaptive enhancement algorithm is proposed based on joint histogram equalization with image fusion. First the noise is suppressed by the use of stationary wavelet decomposition, then the contrast of the image is improved by the adaptive histogram equalization and new high-frequency components are added by the use of nonlinear filter, finally the whole enhancement is realized through image fusion. The disadvantages of traditional technology are coped with, such as the histogram equalization will enhance the background and it is hard to choose the plateau value in plateau histogram equalization algorithm. Experimental results show that it is an effective enhancement algorithm for most IR images, the background is constrained greatly and the contrast of the image is improved significantly. Moreover, the clearness of the target object is also improved significantly.

关 键 词：红外图像增强 对比度增强 自适应阈值分割 直方图均衡 平稳小波变换 

分 类 号：TN215[电子电信—物理电子学]