利用MRF方法的高分辨率影像道路提取
Object-based Road Extraction in Remote Sensing Image Using Markov Random Field
-
摘要: 高分辨率遥感影像提供了地物更丰富的信息,包括光谱信息、地物结构、形状、纹理以及地物之间的空间关系等多方面的信息。面向对象的影像分析方法以目标地物为研究对象,充分考虑目标地物的形状、结构及空间关系等信息进行目标的提取和分析,是当前高分辨率信息提取技术的主要方法。研究了采用面向对象目标的思想将MRF方法应用于高分辨率遥感影像的道路目标提取中,并进行了道路提取实验。Abstract: We study the road extraction method of high resolution image.The method can be defined as a knowledge-based and object-oriented technique.To verify the validity of the method,the road extraction experiments with two types of high resolution remote sensing image are made.The extraction results prove that the algorithm based on MRF model can extract the road information from the high resolution image.