Geometric estimation method of spectral unmixing
基于几何估计的光谱解混方法

spectral unmixing in one of the important techniques of hyperspectral data analysis. Full constrained (i.e. non-negative and sum to one constrainted) least squares linear spectral mixture modeling (FCLS-LSMM) is widely used for its conciseness and clarity of physical meaning. Unfortunately, the traditionally iterative processing for solving FCLS-LSMM is of heavy computational burden. Recently developed geometric analysis method of LSMM provided a new way for lowing down the complexity of LSMM solving. The unmixing results, however, are not in line with the FCLS requirements. In this case, a new geometric unmixing method is constructed to completely meet the FCLS requirements. The method is of very low complexity, at the same time, has the capability to obtain the theoretically optimal solution. Experiments show the effective of the proposed method.