摘要
根据人眼视觉系统特性,分析了场景理解和目标探测对彩色融合的要求,并据此提出了一种通过颜色对比度增强来提高目标可探测性的夜视融合方法.该方法使热目标呈红色,冷目标呈蓝绿色;根据红外图像特征,引入了一种和红外图像各像素亮度与图像平均亮度的偏离相关的颜色对比度增强因子,利用该因子可增强目标与背景的颜色对比度,弥补颜色传递彩色融合方法在颜色对比度上的不足,能有效提高目标可探测性.实验表明该方法既能突出红外目标,又能保持丰富的背景细节,在增强场景理解的同时提高了目标可探测性.
Requirements on color image fusion for situation awareness and target detection tasks were analyzed based on the characteristics of human visual system. Then a night vision fusion method was presented to improve target detectability by enhancing color contrast between the target and the background. The proposed fusion method relates hot and cold targets to the colors of intense red and eyan, respectively. According to the characteristics of infrared images, a color contrast enhancement ratio related to the divergence of the intensity of each pixel from the mean intensity of the infrared image was utilized to enhance color contrast between the target and the background. The enhancing method offsets the disadvantage of color transfer method in color contrast and improves target detectability. Experimental results demonstrate that the method can not only pop out both hot and cold targets, but also keep details of the background well. The situation awareness and target detectability can be improved by the method.
出处
《红外与毫米波学报》
SCIE
EI
CAS
CSCD
北大核心
2009年第4期281-284,共4页
Journal of Infrared and Millimeter Waves
基金
国家自然科学基金(60807005)资助项目
关键词
图像融合
夜视
对比度增强
颜色对比度
目标探测
image fusion
night vision
contrast enhancement
color contrast
target detection
