基于混沌遗传算法的遥感影像分类  

Remote sensing image classification based on Chaos Genetic algorithm

作　　者：黄明[1] 吴延斌[2] 

机构地区：[1]中国矿业大学煤炭资源与安全开采国家重点实验室,北京100083 [2]黑龙江工程学院测绘工程系,哈尔滨150050

出　　处：《测绘科学》2011年第2期5-8,共4页Science of Surveying and Mapping

基　　金：国家"973"项目(2006CB202209);国家自然基金重大项目(50490271)

摘　　要：为提高遥感影像分类精度,本文提出基于混沌遗传算法(Chaos Genetic Algorithm)的遥感影像分类方法。首先应用混沌遗传算法对样本进行自学习得到全局最优的聚类中心,然后通过得到的聚类中心对整幅影像进行分类。该方法利用混沌变量的遍历性,进行粗粒搜索,优化遗传算法的初始种群,从而提高收敛速度;对经过选择算子、交叉算子、变异算子计算得到的优秀个体,利用混沌系统对初始条件和系统参数的敏感性进行混沌扰动,避免陷入局部最优,从而得到全局最优解,获得最优聚类中心。该方法应用于淮南矿区TM影像分类,实验表明该方法分类总正确率为88.26%,Kappa系数为0.853,优于传统分类方法。To improve the accuracy of remote sensing image classification, the Chaos Genetic algorithm was proposed in the paper. The Chaos Genetic algorithm has capability of self-learning, hence through the input samples the global optimization clustering center was found. And then the clustering center was employed to classify the view picture of remote sensing image. In this process, the ergodic property of Chaos phenomenon was used to optimize the initial population, so it could accelerate the convergence of Ge- netic algorithms. Chaotic systems are sensitive to initial condition system parameters. In order to escape from local optimums, the Chaos operator was applied to optimize the individuals after the process of selection operator, mutation operator and crossover operator. Chaos genetic algorithm was applied to classify the TM image of Huainan. Moreover, the classification of the Parallelepiped and Maximum likelihood and Standard Genetic algorithm methods were contrasted with it through confusion matrix. It was demonstrated that Chaos Genetic algorithm was superior to the two traditional algorithms, and its overall accuracy and Kappa coefficient reached 88. 26% and 0. 853 respectively.

关 键 词：遥感影像 影像分类 采煤沉陷区 混沌 遗传算法 

分 类 号：TP751[自动化与计算机技术—检测技术与自动化装置]