首页 | 本学科首页   官方微博 | 高级检索  
     

结合MST划分和RHMRF-FCM算法的高分辨率遥感图像分割
引用本文:林文杰,李玉,赵泉华. 结合MST划分和RHMRF-FCM算法的高分辨率遥感图像分割[J]. 测绘学报, 2019, 48(1): 64-74. DOI: 10.11947/j.AGCS.2019.20170585
作者姓名:林文杰  李玉  赵泉华
作者单位:辽宁工程技术大学测绘与地理科学学院遥感科学与应用研究所,辽宁 阜新,123000;辽宁工程技术大学测绘与地理科学学院遥感科学与应用研究所,辽宁 阜新,123000;辽宁工程技术大学测绘与地理科学学院遥感科学与应用研究所,辽宁 阜新,123000
基金项目:国家自然科学基金(41271435);国家自然科学基金青年科学基金(41301479)
摘    要:针对基于像素的HMRF-FCM算法抗噪性差以及对地物复杂边界分割精度低的问题,提出一种结合形状信息的静态MST区域划分和RHMRF-FCM算法的高分辨率遥感图像分割方法。该方法定义一种静态MST同质区域划分准则,借助MST能较好表达边界和形状信息、能较好抑制几何噪声的特点,解决地物复杂边界的表达和降低分割结果中几何噪声问题。首先,利用MST静态划分将图像域划分成若干个均质区域,假设每个均质区域内光谱测度服从独立同一的多元高斯分布。然后,在此基础上构建了区域隐马尔可夫随机场模型,以及建立基于信息熵和KL信息正则化项的模糊聚类目标函数。最后,采用偏微分方法对分割模型参数进行求解,从而得到全局最优分割结果。为验证本文方法,对WorldView-3高分遥感图像进行分割试验。定性、定量分析了尺度参数、光谱相似性参数和区域紧致度参数对最优分割结果的影响,并对比分析本文算法和eCognition软件中的多分辨率分割算法、分水岭算法。

关 键 词:静态MST划分  形状参数  区域隐马尔可夫随机场  模糊c均值算法  高分辨遥感图像分割
收稿时间:2017-10-16
修稿时间:2018-05-04

High-resolution remote sensing image segmentation using minimum spanning tree tessellation and RHMRF-FCM algorithm
LIN Wenjie,LI Yu,ZHAO Quanhua. High-resolution remote sensing image segmentation using minimum spanning tree tessellation and RHMRF-FCM algorithm[J]. Acta Geodaetica et Cartographica Sinica, 2019, 48(1): 64-74. DOI: 10.11947/j.AGCS.2019.20170585
Authors:LIN Wenjie  LI Yu  ZHAO Quanhua
Affiliation:The Institute of Remote Sensing, School of Geomatics, Liaoning Technical University, Fuxin 123000, China
Abstract:It is proposed that a high-resolution remote sensing image segmentation method that combines static minimum spanning tree tessellation considering shape information and the RHMRF-FCM algorithm. It solves the problems in traditional pixel-based HMRF-FCM algorithm in which poor noise resistance and low precision segmentation in complex boundary exist. By using the MST model and shape information, the object boundary and geometrical noise can be expressed and reduced respectively. Firstly, the static MST tessellation is employed for partitioning the image domain into some polygons corresponded to the components of homogeneous regions needed to be segmented. Secondly, based on the tessellation results, the RHMRF model is built, and regulation term considering the KL information and information entropy are introduced into the FCM objective function. Finally, the partial differential method is employed to calculate the parameters of the fuzzy objective function for obtaining the global optimal segmentation results. To verify the robust and effective of proposed algorithm, the experiments are carried out with WorldView-3 high resolution image. The results from proposed method with different parameters and comparing methods (the multi-resolution and the watershed segmentation method in eCognition software) are analyzed qualitatively and quantitatively.
Keywords:static minimum spanning tree tessellation  shape parameter  regional hidden Markov random field  fuzzy c-means algorithm  high-resolution remote sensing image segmentation
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《测绘学报》浏览原始摘要信息
点击此处可从《测绘学报》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号