摘要
介绍几种常见的分类方法,并对分类结果进行精度评价及比较.结果表明,总体精度基本达到要求,马氏距离法分类结果更理想且保证了精度要求.在具体操作过程中,需要多选择适当的分类方法并进行综合评价,从而选择恰当的分类方法.
Several common classification methods are introduced,and the accuracy of the classification results are evaluated and compared. The results show that the overall accura-cy of the basic requirements, Kappa coefficient in addition to the minimum distance meth-od , mahalanobis distance, the spectral information divergence method can achieve relative-ly good requirements, the overall evaluation results are good. The classification result of the Markov distance method is more ideal and ensures the accuracy of the method. In the process of specific operation, it need to do some classification methods and comprehensive evaluation, so as to select the appropriate classification method.
出处
《牡丹江师范学院学报（自然科学版）》
2017年第1期37-38,共2页
Journal of Mudanjiang Normal University：Natural Sciences Edition
基金
高校自然科学研究项目(KJ2016A770)
高校优秀青年人才支持计划重点项目(gxyqZD2016340)
安徽省大学生创新创业训练计划项目(201510379080
201510379046
201510379084)
宿州学院卓越人才教育培养计划(szxy2015zjjh01)
关键词
遥感
监督分类
分类精度
Remote sensing
Supervised classification
Classification accuracy
