首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于多尺度小波特征的高光谱影像亚像素目标识别
引用本文:路威,余旭初,杨国鹏,马永刚.基于多尺度小波特征的高光谱影像亚像素目标识别[J].海洋测绘,2005,25(2):21-25.
作者姓名:路威  余旭初  杨国鹏  马永刚
作者单位:解放军信息工程大学,测绘学院,河南,郑州,450052;61363部队,陕西,西安,710054
基金项目:国家“863”资助项目(2002AA783050)
摘    要:提出了一种基于多分辨率小波高频特征系数的高光谱遥感影像亚像素目标识别方法。首先利用多尺度小波变换将光谱信号分解为不同尺度的高频特征信号,然后借助接收操作特性曲线(ROC)和马氏距离投影寻踪求取一维最佳识别特征,最后通过高斯最大似然决策函数求解亚像素目标的存在概率。通过38种小波函数的高光谱数据实验证明,该方法对亚像素目标的识别效果较好。

关 键 词:多尺度小波特征  接收操作特性曲线  投影寻踪  高斯最大似然决策函数
文章编号:1671-3044(2005)02-0021-05
修稿时间:2005年1月13日

Subpixel Target Detection Approach Based on Multiscale Wavelet Features in Hyperspectral Imagery
LU Wei,YU Xu-chu,YANG Guo-peng,MA Yong-gang.Subpixel Target Detection Approach Based on Multiscale Wavelet Features in Hyperspectral Imagery[J].Hydrographic Surveying and Charting,2005,25(2):21-25.
Authors:LU Wei  YU Xu-chu  YANG Guo-peng  MA Yong-gang
Institution:LU Wei~1,YU Xu-chu~1,YANG Guo-peng~1,MA Yong-gang~2
Abstract:This paper presents a subpixel target detection approach based on multiscale wavelet high-pass coefficients in hyperspectral data.Firstly this algorithm uses multiscale wavelet transforms to obtain multiscale wavelet high-pass features of spectral signals.Secondly it gets the best-discrimination feature with receiver operating characteristics curves(ROC) and projection pursuit based on mahalanobis distance.Lastly it calculates the probability of subpixel target in every pixel with gauss maximum likelihood decision function.Experiment has been made with 38 kind wavelet functions in hyperspectral imagery,and proved that this algorithm is adapted for detecting subpixel targets and has an excellent precision result.
Keywords:multiscale wavelet features  receiver operating characteristics curves  projection pursuit  gauss maximum likelihood decision function
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号