Abstract: Particle Swarm Optimization（PSO） is an optimization algorithm based on continuous space. Because the number of endmembers is small and endmembers are scattered in the hyperspectral image, the search space of the PSO is scattered. The traditional PSO algorithm has the weaknesses of being sensitive to initial value, low convergency, easy to fall into the local optimum. To solve this problem, this paper selects the pixels with high Pure Pixel Index（PPI） as the preselected pixels, and sorts the preselected pixels. Finally, the sorting pixels are taken as the searching space of PSO, to reduce the search space and improve the efficiency of the algorithm. Experimental results show that this algorithm has better result than other algorithms in simulating and AVIRIS images.

Key words: particle swarm optimization, endmember extraction, hyperspectral remote sensing

摘要： 粒子群优化算法（Particle Swarm Optimization，PSO）应用于高光谱影像端元提取时，由于影像中存在端元的像元数所占比例极小且分布零散，导致粒子群的搜索空间破碎，存在收敛性能低、容易陷入局部最优解等缺陷。对粒子群的搜索空间进行优化，选择影像中纯净像元指数（Pixel Purity Index，PPI）较大的像元作为预选像元，然后对预选像元进行光谱聚类排序，将排序后的集合作为粒子群的搜索空间，优化了粒子的搜索空间。并在迭代过程中，充分利用粒子群的信息自适应地调整其系数，在缩小原始图像与反演图像的误差同时，增加体积约束，在提取端元时更好地保持其原有的形状。通过模拟数据和AVIRIS影像的实验表明该算法具有较好端元提取效果。

关键词: 粒子群算法, 端元提取, 高光谱遥感