基于地物光谱分析的WorldView-2数据岩性识别:以新疆乌鲁克萨依地区为例

Lithology Identification with WorldView-2 Data Based on Spectral Analysis of Surface Features:A Case Study of Wulukesayi District in Xinjiang

针对新疆维吾尔自治区西昆仑乌鲁克萨依地区的岩性出露情况,采用WorldView-2遥感影像数据进行岩性识别研究。在分析研究区内不同岩性的实测光谱曲线特征的基础上,针对在岩性识别过程中出现的不同岩性色彩反差弱、异物同谱、岩层表层浅覆盖等一系列问题,结合高空间分辨率WorldView-2数据波段特点和数据统计特征,确定了最佳波段组合,并采用主成分分析降低了波段间的高相关性。在此基础上进行反差增强、纹理增强等处理,提高了不同岩性对比度,去除了外界因素对光谱信息的干扰,为精确识别不同岩性提供了可靠依据。

Illustrated by the case of Wulukesayi district in Xinjiang Uygur Autonomous Region, we identify lithology by analyzing the WorldView-2 remote sensing image data. On the basis of analyzing the measured spectral curves of different lithologies, aiming at a series of problems arising in the process of lithology identification, such as weak chromatic aberration between different lithologies, shallow cover of the rock surface and different objects with the same spectrum, we determine the optimum band combination according to the band characteristic and statistical features of high spatial resolution WorldView-2 data. The high correlation between bands is reduced by using the method of principal component analysis. On that basis, by conducting contrast enhancement and texture enhancement processing, we improve the contrast between different lithologies and eliminate the interference of external factors on the spectral information, which provides a reliable basis for the precise identification of different lithologies.