CAJ | 학술논문

高光谱遥感通过航空或航天成像光谱仪获得地物的连续光谱信息，可探测到普通遥感不能探测到的物质。而独立成分分析技术是一种基于信号高阶统计特性的分析方法，在利用该方法进行高光谱数据处理时，将待识别的地物作为独立成分表示出来，可得到地物之间的分离度最大化。通过对高光谱遥感图像数据的特性进行分析，使用USGS光谱库中的矿物光谱作为端元信号，构造模拟的高光谱遥感图像，证明了快速定点学习算法在高光谱遥感矿物蚀变信息提取中的有效性。
고광보요감통과항공혹항천성상광보의획득지물적련속광보신식，가탐측도보통요감불능탐측도적물질。이독립성분분석기술시일충기우신호고계통계특성적분석방법，재이용해방법진행고광보수거처리시，장대식별적지물작위독립성분표시출래，가득도지물지간적분리도최대화。통과대고광보요감도상수거적특성진행분석，사용USGS광보고중적광물광보작위단원신호，구조모의적고광보요감도상，증명료쾌속정점학습산법재고광보요감광물식변신식제취중적유효성。
Hyperspectral remote sensing obtains information of continuous spectrum of ground objects by aerial or space imaging spectrometer, detecting the surface features which would not be detected by traditional remote sensing. Currently, geological prospecting is one of the most successful areas of hyperspectral remote sensing application. This paper analyzes the characteristics of hyperspectral remote sensing data, and explores the application of FastlCA technique to hyperspectral image and information extraction of mineral alteration by using USGS Mineral Spectral Library to construct hyperspectral simulated images, thus proving the effectiveness of FastlCA technique in mineral alteration extraction by hyperspectral remote sensing.