基于最小生成树的遥感影像分割

Remote Sensing Image Segmentation Based on Minimum Spanning Tree

随着遥感影像在人们生产生活各个方面的广泛应用,普通的分割算法已不适用于高分辨率遥感影像的目标识别需求。本文提出了一种基于最小生成树的遥感影像分割算法,首先根据像素间的相似性测度构建遥感影像的无向带权图,然后通过Prim算法生成遥感影像的最小生成树用于刻画影像的像素间相似性结构,并建立最小生成树的拓扑结构;在此基础上,建立基于最小生成树的分割模型,进行子树划分,从而达到影像分割的目的。模拟和真实影像的分割实验表明,本文算法是一种有效的遥感影像分割算法。

Along with the wide application of remote sensing image in all aspects of people′s life and production,common segmentation algorithm some does not apply in high resolution remote sensing image target recognition.In this paper,we present a segmentation algorithm of remote sensing image based on the minimum spanning tree,first of all according to the remote sensing image similarity measure between pixels in the construction of undirected weighted graph and the prim algorithm generation for minimum spanning tree of remote sensing image used to portray the image pixel similarity structure,and the establishment of minimum spanning tree topology structure;on the basis of this,a segmentation model based on minimum spanning tree,subtree partition,so as to achieve the purpose of image segmentation.The simulation and Real image segmentation experiments show that the proposed algorithm is an effective method for remote sensing image segmentation.