摘要: 现有的粒子群优化(PSO)算法和遗传算法(GA)无法很好地解决高光谱影像端元提取这类离散解空间内的大规模取样优化问题。针对该问题,借鉴凸面几何学理论,利用局部模式粒子群优化的原理改进遗传算法,提出一种面向高光谱影像端元提取的粒子群优化遗传算法(PSOGA)。利用模拟数据和PHI影像对PSOGA算法和GA算法进行实验对比。分析结果证明,PSOGA算法的收敛速度优于GA算法。
关键词:
高光谱,
粒子群优化算法,
遗传算法,
端元提取,
收敛速度
Abstract: The existing Particle Swarm Optimization(PSO) and Genetic Algorithm(GA) can not solve the optimization problems of sampling in large discrete solution space effectively such as endmember extraction in hyperspectral imagery. The theory of PSO is reviewed. Combined with the convex geometry theory, a Particle Swarm Optimization Genetic Algorithm(PSOGA) for endmember extraction in hyperspectral imagery is proposed, which improves GA with the theory of local best structure of PSO algorithm. It carries out the experiments by simulative and real hyperspectral image, and the results between the PSOGA and GA are compared and analyzed. Experimental results prove the convergence rate of PSOGA is much faster than GA’s.
Key words:
hyperspectral,
Particle Swarm Optimization(PSO) algorithm,
Genetic Algorithm(GA),
endmember extraction,
convergence rate
中图分类号:
陈伟, 余旭初, 张鹏强, 王鹤. 面向端元提取的粒子群优化遗传算法[J]. 计算机工程, 2011, 37(16): 188-190.
CHEN Wei, TU Xu-Chu, ZHANG Feng-Jiang, WANG He. Particle Swarm Optimization Genetic Algorithm for Endmember Extraction[J]. Computer Engineering, 2011, 37(16): 188-190.