一种新的自适应红外图像增强算法  

A New Adaptive Enhancement Algorithm for Infrared Image

作　　者：赖睿[1] 刘上乾[2] 王炳健[2] 李庆[2] 

机构地区：[1]西安电子科技大学微电子学院,陕西西安710071 [2]西安电子科技大学技术物理学院,陕西西安710071

出　　处：《半导体光电》2006年第6期767-769,776,共4页Semiconductor Optoelectronics

基　　金：国家自然科学基金资助项目(60377034)

摘　　要：传统的直方图均衡和灰度变换增强算法,不能针对红外图像的目标进行有效地增强。鉴于此,在研究红外图像特点的基础上,提出了一种基于直方图的自适应红外图像增强算法。该算法通过对原始图像的直方图进行自适应双阈值映射和欠采样投影处理,实现了对图像灰度级的重新分配;在减少冗余灰度级的基础上,有效拓宽了目标部分占据的动态范围,既提高了目标与背景间对比度又突出了目标细节层次。Traditional histogram equalization and gray-scale transformation algorithms can＇t realize the enhancement of the target in the infrared image effectively. Consequently, an adaptive enhancement algorithm based on the study of infrared image histogram has been presented. This algorithm increases the contrast between the target and background and improves the resolution of target details by using adaptive bi-threshold mapping and undersampling projection, which reduces the redundant gray level and broadens the dynamic range of target considerably. The experimental results validate that the algorithm greatly improves the resolution and contrast of infrared image and is simple for computation.

关 键 词：直方图均衡 灰度变换 直方图投影 欠采样投影 欠采样双阈值映射 

分 类 号：TN219[电子电信—物理电子学]