高光谱遥感图像混合像元分解的群智能算法

近年来，通过群智能算法求解组合优化或连续优化问题以实现高光谱图像混合像元分解方面取得了重要进展和显著成果.本文首先回顾了高光谱图像混合像元分解的研究背景和群智能算法的特点，然后梳理了光谱混合模型及对应的最优化模型，进而介绍了基于群智能算法的端元提取和丰度反演方法，最后通过2组实验比较了群智能算法和其他传统算法在端元提取和丰度反演方面的精度，对基于群智能算法的混合像元分解效果进行了评价.另外，本文也对群智能算法在高光谱图像信息提取中应用的优势和存在的问题进行了总结.

Swarm intelligence algorithms for spectral unmixing in hyperspectral image

In recent years,swarm intelligence algorithms have made important progress and remarkable achievements in spectral unmixing of hyperspectral image by solving combinatorial optimization or continuous optimization problems.In this paper,the background of the research of spectral unmixing in hyperspectral image and the characteristics of swarm intelligence algorithm were reviewed firstly,and then the optimization model and the spectral mixture model were teased out correspondingly.Then the endmember extraction and abundance inversion method based on swarm intelligent algorithms were introduced.Finally the accuracy of spectral unmixing achieved by swarm intelligence algorithms and other traditional algorithms was evaluated through two experiments.In addition,the advantages and problems of swarm intelligence algorithm in hyperspectral image information extraction were also summarized in this paper.