Take categorical data matrix and transform whole matrix to binary sparse 1ofM matrix, keeping track of what came where. Ideal for any form of count-based probabilistic analysis.-Take categorical data matrix and transform whole matrix to binary sparse

实现非负矩阵分解算法和独立成分分析，得到遥感遥感图像解混结果(The non negative matrix factorization algorithm and independent component analysis are implemented to get the unmixing results of remote sensing images)