基于Kohonen网络模型,对标准的SOM(self-organizing feature map)算法进行了改进,在保持点群原有空间分布特征的情况下研究点群的选取和典型化。实践表明,该方法适合任意空间分布类型的点群综合。
Selection and typification of points group is discussed based on Kohonen net modal, standard SOM algorithm is improved so as to preserve the spatial distribution properties of original points group. Examples illustrate that this method can be used to generalization of points group with any spatial distribution.