基于差分搜索的高光谱图像解混算法

摘要

针对高光谱图像解混问题进行研究，发现高光谱图像中各个端元的分布不完全独立，不能将盲源分离方法直接应用于高光谱图像解混。为此，提出了一种基于差分搜索的高光谱图像解混算法。该算法根据高光谱图像丰度非负和丰度和为一特性构造相应的约束项，与互信息相结合作为目标函数，利用差分搜索算法对该目标函数进行优化求解来实现高光谱图像解混。仿真数据和实际数据实验表明，该算法能够有效解决高光谱图像解混问题，与已有其他算法相比，能避免陷入局部极值，提高了图像解混的精度，并且针对不含纯像元的高光谱图像具有很好的解混效果。%With regard to the issues of hyperspectral unmixing,the distribution of endmembers were not completely indepen-dent in hyperspectral images,thus could not directly apply blind source separation to hyperspectral unmixing.This paper pro-posed a novel hyperspectral unmixing algorithm based on differential search.According to the abundance non-negative and a-bundance sum-to-one features,this algorithm constructed corresponding constraint terms and combined it with mutual informa-tion as an objective function,and then optimized the function through differential search algorithm to realize hyperspectral un-mixing.The experimental results on simulated and real hyperspectral data demonstrate that the proposed algorithm can effec-tively solve the problem of hyperspectral unmixing.Compared with other algorithms,it can avoid falling into local extremum and get more accurate results,and also be used to unmix hyperspectral data without pure pixels.