基于四叉树与图割的遥感图像分割方法  

Remote Sensing Image Segmentation Approach Based on Quarter-tree and Graph Cut

作　　者：周四龙[1,2] 粱栋[1,2] 王慧[2] 孔颉[2] 

机构地区：[1]安徽大学计算智能与信号处理教育部重点实验室,合肥230039 [2]安徽大学电子科学与技术学院,合肥230039

出　　处：《计算机工程》2010年第8期224-226,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60772121);安徽省自然科学基金资助项目(070412065);安徽省高等学校自然科学研究基金资助项目(2005KJ005ZD);安徽大学“211”工程学术创新团队基金资助项目

摘　　要：针对图割算法实时性差的问题,提出一种基于四叉树与图割的遥感图像分割方法。用四叉树分割方法将遥感图像分割成多个区域,根据区域的邻接性合并相似小区域,结合区域间的灰度和位置信息,利用Normalized Cut方法在区域间进行划分,完成图像的分割。实验结果证明该方法能有效消除四叉树方法的过分割现象。Aiming at the problem of poor real-time ability of graph cut,this paper suggests a remote sensing image segmentation approach based on quarter-tree and graph cut.The approach uses quarter-tree to segment the remote sensing image into a large number of small partitions,with the adjacency of each partition,smaller and similar partitions are merged into some areas.By integrating gray feature and spatial location of each partition,normalized cutting approach is used to segment the image among partition from global view,by which the final segmented image can be generated.Experimental results show that the over-segmentation can be eliminated effectively.

关 键 词：图像分割 四叉树 区域合并 图割 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]