论文部分内容阅读
端元提取技术是混合像元分解中重要的步骤之一,传统的端元提取方法仅考虑了像元的光谱信息。本文将数学形态学算子扩展到高光谱空间,并应用到端元提取技术中,可以顾及像元的上下文信息。利用AVIRIS高光谱仿真数据对算法进行了实验验证,结果表明本文算法具有较强的抗噪能力和较高的可靠性。在此基础上,结合徐州地区的EO-1 Hyperion高光谱遥感图像,使用本文算法进行了端元提取应用研究,将实验结果与纯净像元指数、顶点成分分析方法做了对比分析和精度评价,证明本文算法是一种可靠的高光谱遥感图像端元提取技术。
Endmember extraction is one of the most important steps in mixed pixel decomposition. Traditional endmember extraction only considers the spectral information of pixels. In this paper, the mathematical morphological operator is extended to the hyperspectral space and applied to the endmember extraction technology, which can take into account the contextual information of the pixel. The experimental results show that the proposed algorithm has strong anti-noise ability and high reliability by using AVIRIS hyper-spectral simulation data. On this basis, combined with the EO-1 Hyperion hyperspectral remote sensing image in Xuzhou area, the end-element extraction application is studied by using the algorithm in this paper. The experimental results are compared with the pure pixel index and vertex component analysis method, Prove this algorithm is a reliable end hyperspectral remote sensing image extraction technology.