高分辨率遥感影像分割方法研究

在遥感应用分析中,遥感影像分割是低层影像处理和中高层影像分析和理解的桥梁,是实现遥感影像信息自动提取的关键步骤,具有重要的意义。随着大量高分辨率遥感影像的出现,传统基于像素的影像处理方法已不能适应高分辨率遥感影像。近年来,国内外研究者们提出了面向对象影像的分析方法,而面向对象影像分析方法的关键就是影像分割,影像分割精度直接影响着高分辨率遥感信息提取和目标识别的精度。首先给出一般图像分割方法的综述;然后分析和总结了当前主要的高分辨率遥感影像分割方法,着重阐述了均值漂移、分形网络进化、马尔科夫随机场等分割方法的特点和研究现状;最后,对高分辨率遥感应用分析中影像分割方法的发展趋势进行了讨论与展望。

Study on Segmentation Methods of High Resolution Remote Sensing Images

In remote sensing application analysis,image segmentation is the bridge between low- level image processing and high-level image analysis and understanding. It is also the first step to realize automatic extraction from remote sensing images and is of great significance. As a large number of high resolution remote sensing images are obtained,traditional image segmentation methods are not meeting the rising demands of high resolution remote sensing image analysis. In recent years,object- based image analysis（ OBIA） method is proposed by domestic and foreign researchers. The key of OBIA is image segmentation which directly influences the accuracy of high resolution remote sensing information extraction and target recognition. In this paper,a review of general image segmentation methods are firstly given,then remote sensing image segmentation methods which are widely used at present are summarized and analyzed. Characteristics and research status of methods such as Mean Shift algorithm,FNEA,MRF etc. are focused on. In the end,the development tendency of image segmentation in high resolution remote sensing application analysis is discussed and prospected.