基于改进的Mean Shift方法的高分辨率遥感影像道路提取

遥感影像的道路提取长期以来一直是数字摄影测量未能解决的难题之一。本文在探讨了均值漂移、区域分割、边缘检测、轮廓跟踪的基础上,提出一种基于均值漂移和利用统计面积去除和合并小区域的道路提取算法,并采用数学形态学方法消除错误道路,利用轮廓跟踪法获取道路的边缘,实现道路的提取;通过与基于区域分割、边缘检测算法提取结果对比,结果表明,本文的算法进行道路提取的效果更好。

Road Extraction from High Resolution Remote Sensing Images Based on the Improved Mean Shift Algorithm Method

Road extraction from remote sensing image has been one of the unsolved problems for a long time in digital photogrammetry.Taking the mean shift,regional segment,edge detection,contour tracking as the foundation,this paper puts forward a road extraction algorithm based on mean shift and using statistical areas to remove and merge small regions,and uses mathematical morphology to eliminate wrong road,and then acquires the edge by contour tracking,thus realizes the road extraction.Comparing with the results of regional segmentation and edge detection,the results show that the algorithm proposed in this paper can get a best effect.