基于灰度空间相关性的双谱微光图像融合方法  

Method of fusing dual-spectrum low light Level images based on gray-scale spatial correlation

作　　者：张闯[1] 柏连发[1] 张毅[1] 

机构地区：[1]南京理工大学电光学院,南京210094

出　　处：《物理学报》2007年第6期3227-3233,共7页Acta Physica Sinica

基　　金：国家自然科学基金(批准号:60572107)资助的课题~~

摘　　要：图像的灰度空间相关性可以反映图像的清晰度,而图像融合的主要目的之一就是改善图像的清晰度.根据微光全波图像和微光短波图像的光谱特点,在分析微光全波图像和微光短波图像的一维灰度直方图及二维灰度空间相关性图的基础上,提出了一种新型的基于灰度空间相关性的图像融合方法.该融合方法由基于标准偏差的灰度调制和灰度统计平衡两部分实现,同灰度调制融合法及谱域融合法比较,此方法能够有效地改善图像的清晰度,同时便于硬件实现.文中详细阐述了该融合方法的理论公式,并分析了其在不同场景时的实验结果.Gray-scale spatial correlation reflects the definition of images, and the main purpose of fusing images is to improve the definition of images. Based on the spectrum characteristics of low light level （LEL） whole-wave image and LLL short-wave image and the analysis of the one-dimension at gray-scale histograms and the two-dimensional gray-scale spatial correlation charts of LLL whole-wave image and LLL short-wave image, the new method of fusing dual-spectrum LLL images based on gray-scale spatial correlation is proposed. The method is realized by gray-scale selection based standard deviation and gray-scale statistics balance, and the method can effectively improve the definition of images and is easy to execute on hardware compared with gray-scale modulation method and spectrum-tield method. The theoretical expressions of the fusion method are expatiated particularly, and the experiment results with different scenes are analyzed.

关 键 词：图像融合 微光全波图像 微光短波图像 灰度空间相关性 

分 类 号：TN911.73[电子电信—通信与信息系统]