光子学报, 2009, 38 (12): 3165, 网络出版: 2010-05-10   

基于新型光谱相似度量核的高光谱异常检测算法

A Novel Spectral Similarity Measurement Kernel Based Anomaly Detection Method in Hyperspectral Imagery
作者单位
哈尔滨工程大学 信息与通信工程学院,哈尔滨 150001
摘要
提出一种新型光谱相似度量核函数,并应用于高光谱异常检测.由于高斯径向基核函数是基于光谱向量间欧式距离的度量,其对于光谱向量间距离变化的适应性较强,而对于因光照强度变化,阴影和遮挡等引起的同种地物光谱变化的适应性较弱,使得基于高斯径向基核的高光谱异常检测算法性能下降.为解决该问题,从光谱曲线形状描述出发,基于光谱相似度量提出了光谱相似度量核函数.通过理论分析和真实高光谱数据异常检测实验检验,实验结果说明相对于高斯径向基核函数,光谱相似度量核函数具有一定的优越性,能改善基于核方法的高光谱异常检测算法的性能.
Abstract
A novel spectral similarity measurement kernel function is proposed and applied to anomaly detection in hyperspectral imagery.As the Gaussian Radial Basis Function (RBF) is based on the Euclidean distance of two spectral vectors,it is sensitive for distance variations of two spectral vectors,but not for spectral curve variation coming from radiation intensity variation,shadow,and shading etc.When the spectral curves of a material are variety,the detection performance of the RBF based anomaly detectors degenerate.In order to solve the spectral curves variation problems for the same materials,a spectral similarity measurement kernel function is proposed according to the spectral curves similarity description.A theoretical analysis is expounded and numerical experiments are conducted on real hyperspectral imagery.The detection result comparison of Gaussian Radial Basis Function based and Spectral Similarity Measurement Kernel based anomaly detector shows the Spectral Similarity Measurement kernel can improve the performance of kernel base anomaly detection methods in hyperspectral imagery.

梅锋, 赵春晖, 孙岩, 王立国. 基于新型光谱相似度量核的高光谱异常检测算法[J]. 光子学报, 2009, 38(12): 3165. MEI Feng, ZHAO Chun-hui, SUN Yan, WANG Li-guo. A Novel Spectral Similarity Measurement Kernel Based Anomaly Detection Method in Hyperspectral Imagery[J]. ACTA PHOTONICA SINICA, 2009, 38(12): 3165.

本文已被 2 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!