论文部分内容阅读
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。
In order to improve the classification accuracy of remote sensing images, a remote sensing image classification method based on chaos genetic algorithm is proposed in this paper. Firstly, chaos genetic algorithm is used to self-study the samples to get the global optimal cluster center, and then the whole image is classified by the obtained cluster centers. The method exploits the ergodicity of chaos variables, performs coarse-grained search, and optimizes the initial population of genetic algorithm, so as to improve the convergence rate. For the excellent individuals calculated through selection operator, crossover operator and mutation operator, Conditions and the sensitivity of system parameters to chaotic perturbation, to avoid falling into the local optimum, so as to obtain the global optimal solution and obtain the optimal cluster center. The method is applied to the classification of TM images in Huainan mining area. Experiments show that the method has a total classification accuracy of 88.26% and a Kappa coefficient of 0.853, which is superior to the traditional classification methods.