设计完成和比较了自组织网络的几种算法在遥感土地覆盖分类中的应用，结果表明非监督和监督学习相结合方法进行遥感土地覆盖分类，各算法在分类性能上无显著差异，因此可采用算法较简单的简单竞争学习网络，根据最邻近原则进行非参数分类 This paper designs and compares the application of several algorithms of self-organizing network in remote sensing land cover classification. The results show that the combination of unsupervised and supervised learning can be used to carry out remote sensing land cover classification. There is no significant difference in the classification performance of each algorithm, The algorithm is simpler and simple competitive learning network, according to the nearest neighbor principle of non-parametric classification