摘要： 粒子群优化算法是基于群智能的随机全局优化技术，将它引入航空影像纹理分类，在提取纹理样本小波、分形等特征的基础上，提出了针对分类问题的粒子表达方法和群体寻优策略，实现了基于粒子群算法的纹理分类。将其与基于遗传算法的纹理分类法作比较，结果表明粒子群优化算法具有较好的寻优性能，基于该算法的纹理分类法分类精度较高且计算时间较少。

关键词: 粒子群优化算法, 纹理分类, 航空影像, 特征提取, 遗传算法

Abstract: Particle Swarm Optimization(PSO) is a stochastic global optimization technique based on swarm intelligence. This paper introduces aerial image texture classification. The particles in the swarm are constructed and swarm search strategies are proposed in terms of the needs of classification application after many texture features such as wavelet features and factual features extracted from texture samples. The classification method as PSO is implemented. Compared with the texture classification method based on Genetic Algorithms(GA), PSO has better search ability and the classification method based on it has higher classification accuracy and needs less computation time in the experiment.

Key words: Particle Swarm Optimization(PSO), texture classification, aerial image, feature extraction, Genetic Algorithms(GA)

中图分类号: 

• TP391.41