克隆选择算法在遥感影像分类中的应用

 针对遥感影像分类中的局部极值、鲁棒性等问题，提出基于克隆选择算法的遥感影像监督分类方法。所提方法将遥感影像各波段亮度值定义为抗原的属性，通过计算遥感像元与抗体的亲和力识别其类别，并采用实数制方式对抗体进行变异，在保证亲和力上升的同时，也保证了系统的多样性。该算法应用于广州市遥感影像数据分类的实验结果显示：在抗体的进化过程中，抗体的亲和力和识别能力不断提高，最终的分类精度达到92.9%。与最大似然法相比，克隆选择算法的分类精度更高。

Clonal Selection Algorithm for Classification of Remote Sensing Imagery

A novel supervised classifier for remote sensing data by employing clonal selection algorithm(CSA)is presented for solving the problems,eg.local optimum and robustness,in remote sensing imagery classification.In the classifier,band brightness is defined as antigens attribute,and the image is classified into class with the maximum affinity by calculating the affinity between remote sensing pixel and antibody.Antibody's real encoding mutation integrates affinitys ascending and systems variety. The concerned experiment shows that antibodys affinity increases along with its evolution, and CSA classifications overall accuracy is 92.9%. Comparing with conventional Maximum Likelihood, CSA can get better precision.