基于高光谱辐亮度图像进行目标探测的大气状态影响分析  

Impact Analysis of Atmospheric State for Target Detection in Hyperspectral Radiance Image

作　　者：张兵[1] 沙建军[1,2] 王向伟[1] 高连如[1] 

机构地区：[1]中国科学院对地观测与数字地球科学中心,北京100094 [2]中国科学院研究生院,北京100049

出　　处：《光谱学与光谱分析》2012年第8期2043-2049,共7页Spectroscopy and Spectral Analysis

基　　金：国家(863)计划项目(2011AA120203);国家自然科学基金项目(40901225;41101394)资助

摘　　要：基于高光谱辐亮度图像直接进行目标探测可以提高数据处理效率,满足实时处理的要求。然而遥感器所获取的辐亮度光谱信息会受到大气的影响。通过对高光谱成像过程的模拟,分析大气状态变化对辐亮度图像中目标探测性能的影响。研究结果表明:高光谱辐亮度图像可以直接用于目标探测,不同的大气状态对RXD异常探测影响很小,而MF探测则受输入光谱准确度的影响,待检测目标的辐亮度光谱要求与辐亮度图像获取时的大气状态一致,才有良好的效果。Target detection based on hyperspectral radiance images can improve data processing efficiency to meet the requirements of real-time processing. However, the spectral radiance acquired by the remote sensor will be affected by the atmosphere. In the present paper, hyperspectral imaging process is simulated to analyze the effects of the changes in atmospheric state on tar-get detection in hyperspectral radiance image. The results show that hyperspectral radiance image can be directly used for target detection, different atmospheric states have little impacts on the RXD detection, whereas the MF detection is dependent on the accuracy of the input spectrum, and good results can only be obtained by the MF detector when the atmospheric states are similar between the radiance spectrum of the target to be detected and the simulated hyperspeetral image.

关 键 词：辐亮度图像 大气状态 光谱差异 目标探测 

分 类 号：TP722[自动化与计算机技术—检测技术与自动化装置]