基于动力演化算法的投影寻踪在高光谱影像异常目标检测中的应用  

Application of Projection Pursuit Based on Dynamical Evolutionary Algorithm to Anomaly Target Detection in Hyperspectral Images

作　　者：陈蜜[1] 易尧华[1] 李德仁[1] 秦前清[1] 

机构地区：[1]武汉大学测绘遥感信息工程国家重点实验室,武汉市珞喻路129号430079

出　　处：《武汉大学学报（信息科学版）》2006年第1期55-58,共4页Geomatics and Information Science of Wuhan University

基　　金：国家自然科学基金资助项目(40204008)

摘　　要：运用基于波段间相关性的高光谱影像波段选取方法进行波段的预选取,采用投影寻踪的方法在动力演化算法的基础上寻找最佳投影方向,将高维数据投影至低维数据空间,在各投影分量图像上采用零点检测阈值化的方法进行异常目标的提取。实验结果表明了基于动力演化算法的投影寻踪在高光谱影像异常目标检测中的有效性。This paper introduces the methods of hyperspectral images band selection based on the property of Hyperspectral remote sensing images, utilizes the Projection Pursuit approach to find optimal solutions for a selected projection index based on dynamical evolutionary algorithm and then project a high dimensional data set into a low dimensional data space to produce a sequence of projection images,explores zero-detection method to threshold projection images to detect anomaly target. The experimental results show that the Projection Pursuit based on dynamical evolutionary approach is an effective means to detect anomaly target in hyperspectral images.

关 键 词：高光谱影像 投影寻踪 动力演化算法 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]