空间4-邻域稀疏表示的高光谱图像目标检测  

Hyperspectral imaging target detection algorithm based on spatial 4 neighborhoods for sparse representation

作　　者：赵春晖[1] 李晓慧[1] 朱海峰[1] 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2013年第9期1171-1178,共8页Journal of Harbin Engineering University

基　　金：国家自然科学基金资助项目(61077079);教育部博士点计划基金资助项目(20102304110013);黑龙江省自然科学重点基金资助项目(ZD201216);哈尔滨市优秀学科带头人基金资助项目(RC2013XK009003)

摘　　要：稀疏表示方法在高光谱图像目标检测中取得了较好的检测效果,但其只利用了图像的光谱信息,没有考虑空间信息.针对高光谱图像的空间相关性对目标检测算法的影响,提出了分别采用4-邻域平滑稀疏模型和4-邻域联合稀疏模型对高光谱图像进行目标检测的算法,将目标像素及其4-邻域像素的稀疏表示综合考虑,提高检测算法的效果和效率.使用3组高光谱图像数据进行了仿真实验,实验结果表明,所提出的2种方法分别在检测效果和计算效率上有一定程度的提高.Recent studies indicate sparse representation has achieved very good results in hyperspectral imaging tar-get detections. However, sparse representation only uses the spectral information of the hyperspectral imaging, with-out considering the spatial information. The purpose of this paper is to research the influence of hyperspectral ima- ging＇s spatial correlation on the target detection algorithm, and propose two new sparse representation models, which named 4-neighborhood smoothing sparse model and 4-neighborhood union sparse model, considering sparse representation of both target pixels and its 4-neighborhood pixels to improve the effectiveness and efficiency of the target detection algorithm. With three real hyperspectral imaging datasets in the simulation experiment, the results show that the algorithms utilizing the new models could improve the effectiveness or efficiency of the test results to a certain degree.

关 键 词：高光谱图像 目标检测 稀疏表示 4-邻域 

分 类 号：TP751.1[自动化与计算机技术—检测技术与自动化装置]