| 注册
首页|期刊导航|国土资源遥感|基于ETM+的遥感蚀变信息提取方法对比研究

基于ETM+的遥感蚀变信息提取方法对比研究

张楠楠 周可法 陈曦 李宏

国土资源遥感Issue(2):34-40,7.
国土资源遥感Issue(2):34-40,7.DOI:10.6046/gtzyyg.2012.02.07

基于ETM+的遥感蚀变信息提取方法对比研究

A Comparative Study of Extraction Methods for Alteration Information Based on ETM +

张楠楠 1周可法 2陈曦 1李宏3

作者信息

  • 1. 中国科学院新疆生态与地理研究所,乌鲁木齐830011
  • 2. 新疆维吾尔自治区遥感中心,乌鲁木齐830011
  • 3. 中国科学院新疆矿产资源研究中心,乌鲁木齐830011
  • 折叠

摘要

Abstract

In order to understand, grasp and apply the characteristics of altered wall rock and the original rock from microcosmic and macroscopic angles and look for large deposits or large ore districts, the authors summarized the common extraction models of alteration information,combined the principal component analysis (PCA) with support vector machines ( SVM) to establish the MASK process - PCA - SVM ( MPS ) model, applied and comparatively studied these models in Hatu area of Xinjiang,and made use of the two methods to verify the extraction results. The results show that the extracting results of the common models are rather poorly consistent with the known deposits, but the alteration information extracted by MPS model is better concordant with the known mineral deposits and altered belts, with the coincidence rate arriving at 86.51%. The combination of geological experts'knowledge with geological maps and a known anomaly spot reveals that the result of MPS model is in accord with the geological fact. Therefore, the extracting precision based on MPS model is higher than that of the common models, and the effect of application is fairly good in the study area. The model proposed in this paper provides a new idea for extracting remote sensing alteration information in the future.

关键词

ETM+/蚀变/信息提取/SVM

Key words

ETM/alteration/information extraction/SVM

分类

信息技术与安全科学

引用本文复制引用

张楠楠,周可法,陈曦,李宏..基于ETM+的遥感蚀变信息提取方法对比研究[J].国土资源遥感,2012,(2):34-40,7.

基金项目

中国科学院数字地球重点实验室开放基金项目(编号:2011LDE005)、国家重点基础研究发展计划(编号:2007CB411308)、国家科技支撑计划(编号:2006BAB07B07-04)、新疆自然科学基金项目(编号:200821185)和新疆科技厅项目(编号:200815116)共同资助. (编号:2011LDE005)

国土资源遥感

OA北大核心CSCDCSTPCD

2097-034X

访问量0
|
下载量0
段落导航相关论文