基于光谱特征混合分析的高光谱遥感数据挖掘技术  

Hyperspectral Remote Sensing Data Mining Technology Based on Spectral Feature Mixed Analysis

作　　者：郑丰收 陶为翔 李京 ZHENG Fengshou;TAO Weixiang;LI Jing(Zhengyuan Geographic Information Group Co. Ltd., Beijing 101300, China)

机构地区：[1]正元地理信息集团股份有限公司,北京101300

出　　处：《微型电脑应用》2021年第11期196-199,共4页Microcomputer Applications

摘　　要：为了实现高光谱遥感检测,提高地物目标的遥感检测能力,提出基于光谱特征混合分析的高光谱遥感数据挖掘技术。通过模糊度检测方法进行高光谱遥感数据信息检测,提取高光谱遥感数据的自相关统计特征量,并采用差异度模糊信息挖掘方法进行高光谱遥感数据的谱分析。通过信息融合方法进行高光谱遥感数据的光谱滤波处理,建立高光谱遥感数据特征提取散射模型。最后,建立高光谱遥感数据的光谱信息增强模型,通过分析高光谱遥感数据的光谱特征量进行特征提取和分类识别,实现高光谱遥感数据的优化挖掘。仿真结果表明,采用该方法进行高光谱遥感数据挖掘的精度较高,提高了高光谱遥感数据的检测和谱特征提取能力,从而提高了地物遥感目标的检测识别能力。In order to realize hyperspectral remote sensing detection and improve the detection ability of ground objects,hyperspectral remote sensing data mining technology based on spectral feature hybrid analysis is proposed.The method of fuzzy degree detection is used to detect the information of hyperspectral remote sensing data,and the autocorrelation statistical feature of hyperspectral remote sensing data are extracted,and the spectrum analysis of hyperspectral remote sensing data is carried out by using the difference degree fuzzy information mining method.The spectral filtering of hyperspectral remote sensing data is carried out by information fusion method,and the scattering model of hyperspectral remote sensing data feature extraction is established.Finally,the spectral information enhancement model of hyperspectral remote sensing data is established.By analyzing the spectral feature of hyperspectral remote sensing data,feature extraction and classification recognition are carried out to realize the optimization of hyperspectral remote sensing data mining.The simulation results show that the accuracy of hyperspectral remote sensing data mining is high,the detection and spectral feature extraction capabilities of hyperspectral remote sensing data are improved,and the detection and recognition ability of ground objects is improved.

关 键 词：光谱特征 混合分析 高光谱遥感数据 特征提取 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]