基于多阈值分割的红外图像伪彩增强算法  

Pseudo-color Enhancement Algorithm of Infrared Images Based on Multi-Threshold Segmentation

作　　者：代少升[1] 申娇 陈桂芳[1] 唐华明[1] 王飞[1] 

机构地区：[1]I重庆邮电大学信号与信息处理重庆市重点实验室,重庆400065

出　　处：《半导体光电》2015年第5期800-803,共4页Semiconductor Optoelectronics

基　　金：国家自然科学基金项目(61275099)

摘　　要：针对现有图像增强中的彩虹编码采用固定阈值分割方法,导致增强后的图像缺乏层次感、视觉效果模糊的问题,提出一种基于多阈值分割的红外图像伪彩色增强算法。该算法通过建立伪彩映射模型,采用多阈值分割手段,将不同灰度值映射成丰富的颜色值,以突出图像边缘的细节信息。实验结果表明,该算法处理后的红外图像具有目标轮廓清晰、视觉效果理想的优点。Traditional rainbow coding in existing image enhancement system with fixed threshold segmentation presents problems of lacking sense of hierarchy and fuzzy visual effect.Hence it was proposed one infrared image pseudo-color enhancement algorithm based on multithreshold segmentation.Through building pseudo-color mapping model,the algorithm maps grey values into rich color values by means of multi-threshold segmentation in order to highlight the edge details of the image.The results suggest that the infrared image has clear target contour and ideal visual effect after the process of the algorithm.

关 键 词：红外图像 彩虹编码 最大类间方差法 多阈值 

分 类 号：TN219[电子电信—物理电子学]