正在加载图片...
第9卷第3期 智能系统学报 Vol.9 No.3 2014年6月 CAAI Transactions on Intelligent Systems Jun.2014 D0:10.3969/j.issn.1673-4785.201309081 网络出版地址:http://www.enki..net/kcms/doi/10.3969/j.issn.1673-4785.201309081.html 基于主成分建模的SVDD高光谱图像异常检测 曾现灵,张立燕2,胡荣华 (1.首都师范大学资源环境与地理信息系统北京市重点实验室,北京100048:2.首都师范大学三雏信息获取与应 用教育部重点实验室,北京100048) 摘要:针对SVDD背景建模时混入异常点造成的检测率下降的问题,提出了基于主成分建模的SVDD方法并应用 于高光谱图像异常检测。利用高光谱图像的光谱特征提取背景的主要成分,并分别对不同成分构建超球体,形成单 种背景成分SVDD模型,最后利用综合决策函数对单个SVDD背景模型进行综合判断待检测像元,从而实现高光谱 图像异常像元的检测。用仿真数据和真实数据对算法的性能进行验证,并将其与SVDD方法进行性能比较。结果表 明,新算法在低虚警概率下较之SVDD模型有更高的检测概率,实验结果证明了算法的有效性。 关键词:主成分建模:SVDD:局部邻域聚类;光谱角余弦:高光谱异常检测 中图分类号:TP751文献标志码:A文章编号:1673-4785(2014)03-0343-06 中文引用格式:曾现灵,张立燕,胡荣华.基于主成分建模的SVDD高光谱图像异常检测[J].智能系统学报,2014,9(3):343-348. 英文引用格式:ZENG Xianling,ZHANG Liyan,HU Ronghua.An SVDD algorithm for hyperspectral anomaly detection based on principal component modeling[J].CAAI Transactions on Intelligent Systems,2014,9(3):343-348. An SVDD algorithm for hyperspectral anomaly detection based on principal component modeling ZENG Xianling',ZHANG Liyan2,HU Ronghua' (1.Key Laboratory of Resource Environment and GIS of Beijing,Capital Normal University,Beijing 100048,China;2.Key Laboratory of 3D Information Acquisition and Application of the Ministry of Education,Capital Normal University,Beijing 100048,China) Abstract:An SVDD algorithm based on the principal component modeling is presented for hyperspectral anomaly detection,in order to solve the problem of its low detection rate caused by mixing abnormal points in the process of modeling background.This method extracts the principal components of the background samples by using the hyper- spectral image's spectral signature,and then uses these different components to build different super spheres re- spectively,forms different single background component SVDD models by these super spheres,finally uses the inte- grated decision function to judge these SVDD background models to detect any anomalies.The performance of the algorithm is verified by simulated and real data.The results show that the proposed method can obtain a higher de- tection rate under low false rate than the algorithm based on SVDD,verifying the effectiveness of this proposed method. Keywords:principal component modeling;SVDD;local neighborhood clustering;spectral angle cosine;hyper- spectral anomaly detection 高光谱图像具有图谱合一的性质,其连续的光 谱曲线可以反映被检测对象的物理特征,在复杂的 环境下可用来进行目标检测。高光谱异常检测技术 收稿日期:2013-09-27.网络出版日期:2014-06-17 的应用价值不断凸显,并且近年来成为了高光谱图 基金项目:国家自然科学基金资助项目(41201075):北京市教委科技资 助项目(KM201210028012). 像研究的热点。 通信作者:张立燕.E-mail:zhangliyan010@126.com 经典的高光谱异常检测算法主要有:1990年第 9 卷第 3 期 智 能 系 统 学 报 Vol.9 №.3 2014 年 6 月 CAAI Transactions on Intelligent Systems Jun. 2014 DOI:10.3969 / j.issn.1673⁃4785.201309081 网络出版地址:http: / / www.cnki.net / kcms/ doi / 10.3969 / j.issn.1673-4785.201309081.html 基于主成分建模的 SVDD 高光谱图像异常检测 曾现灵1 ,张立燕2 ,胡荣华1 (1. 首都师范大学 资源环境与地理信息系统北京市重点实验室,北京 100048; 2. 首都师范大学 三维信息获取与应 用教育部重点实验室,北京 100048) 摘 要:针对 SVDD 背景建模时混入异常点造成的检测率下降的问题,提出了基于主成分建模的 SVDD 方法并应用 于高光谱图像异常检测。 利用高光谱图像的光谱特征提取背景的主要成分,并分别对不同成分构建超球体,形成单 种背景成分 SVDD 模型,最后利用综合决策函数对单个 SVDD 背景模型进行综合判断待检测像元,从而实现高光谱 图像异常像元的检测。 用仿真数据和真实数据对算法的性能进行验证,并将其与 SVDD 方法进行性能比较。 结果表 明,新算法在低虚警概率下较之 SVDD 模型有更高的检测概率,实验结果证明了算法的有效性。 关键词:主成分建模;SVDD;局部邻域聚类;光谱角余弦;高光谱异常检测 中图分类号: TP751 文献标志码:A 文章编号:1673⁃4785(2014)03⁃0343⁃06 中文引用格式:曾现灵,张立燕,胡荣华. 基于主成分建模的 SVDD 高光谱图像异常检测[J]. 智能系统学报, 2014, 9(3): 343⁃348. 英文引用格式:ZENG Xianling, ZHANG Liyan, HU Ronghua. An SVDD algorithm for hyperspectral anomaly detection based on principal component modeling[J]. CAAI Transactions on Intelligent Systems, 2014, 9(3): 343⁃348. An SVDD algorithm for hyperspectral anomaly detection based on principal component modeling ZENG Xianling 1 , ZHANG Liyan 2 , HU Ronghua 1 (1. Key Laboratory of Resource Environment and GIS of Beijing, Capital Normal University, Beijing 100048,China; 2. Key Laboratory of 3D Information Acquisition and Application of the Ministry of Education, Capital Normal University, Beijing 100048,China) Abstract:An SVDD algorithm based on the principal component modeling is presented for hyperspectral anomaly detection, in order to solve the problem of its low detection rate caused by mixing abnormal points in the process of modeling background. This method extracts the principal components of the background samples by using the hyper⁃ spectral image’s spectral signature, and then uses these different components to build different super spheres re⁃ spectively, forms different single background component SVDD models by these super spheres, finally uses the inte⁃ grated decision function to judge these SVDD background models to detect any anomalies. The performance of the algorithm is verified by simulated and real data. The results show that the proposed method can obtain a higher de⁃ tection rate under low false rate than the algorithm based on SVDD, verifying the effectiveness of this proposed method. Keywords:principal component modeling; SVDD; local neighborhood clustering; spectral angle cosine; hyper⁃ spectral anomaly detection 收稿日期:2013⁃09⁃27. 网络出版日期:2014⁃06⁃17 . 基金项目:国家自然科学基金资助项目(41201075);北京市教委科技资 助项目(KM201210028012). 通信作者:张立燕.E⁃mail:zhangliyan010@ 126.com. 高光谱图像具有图谱合一的性质,其连续的光 谱曲线可以反映被检测对象的物理特征,在复杂的 环境下可用来进行目标检测。 高光谱异常检测技术 的应用价值不断凸显,并且近年来成为了高光谱图 像研究的热点。 经典的高光谱异常检测算法主要有:1990 年
向下翻页>>
©2008-现在 cucdc.com 高等教育资讯网 版权所有