首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于Radarsat-1数据的北极海冰SAR图像自动分割研究
引用本文:赵庆平,于力刚,王耀,张馨月.基于Radarsat-1数据的北极海冰SAR图像自动分割研究[J].唐山学院学报,2018,31(6):13-17.
作者姓名:赵庆平  于力刚  王耀  张馨月
作者单位:淮北师范大学 物理与电子信息学院, 安徽 淮北 235000,淮北师范大学 实验室与设备管理处, 安徽 淮北 235000,塔斯马尼亚大学 信息技术与系统学院, 澳大利亚 塔斯马尼亚 7005,塔斯马尼亚大学 信息技术与系统学院, 澳大利亚 塔斯马尼亚 7005
基金项目:国家自然科学基金青年项目(6120327);安徽省高等学校自然科学研究项目(KJ2016A650,KJ2016A628)
摘    要:基于Radarsat-1数据提出了一种集成Gibbs采样标记步骤的北极海冰SAR图像分割算法,该算法综合考虑了斑点噪声、纹理特征等不确定因素,经由像素到区域再到大尺度区域这一途径,把区域化标识、Gibbs采样标记以及区域合并等操作组合起来,实现对海冰SAR图像的分割,针对S.Laurence Bay和Beaufort海冰图像的测试实验表明,与基于区域的MRF分割算法相比提出的算法可有效提高图像分割的准确性。

关 键 词:SAR海冰图像  图像分割  Radarsat-1数据  Gibbs采样标记

Automatic SAR Image Segmentation of Arctic Sea Ice Based on Radarsat 1 Data
ZHAO Qing-ping,YU Li-gang,WANG Yao and ZHANG Xin-yue.Automatic SAR Image Segmentation of Arctic Sea Ice Based on Radarsat 1 Data[J].Journal of Tangshan College,2018,31(6):13-17.
Authors:ZHAO Qing-ping  YU Li-gang  WANG Yao and ZHANG Xin-yue
Institution:School of Physics and Electronic Information, Huaibei Normal University, Huaibei 235000, China,Department of Laboratory and Equipment, Huaibei Normal University, Huaibei 235000, China,School of Information Technology and System, University of Tasmania, Tasmania 7005, Australia and School of Information Technology and System, University of Tasmania, Tasmania 7005, Australia
Abstract:Based on Radarsat-1 data, the authors propose a SAR image segmentation algorithm for Arctic sea ice that integrates Gibbs sampling steps. This method considers the uncertainties of speckle noise and texture features, and combines the regionalization mark, Gibbs sampling,and region merging together via pixel, region and larger-scale region. It successfully segmented the SAR image of Arctic sea ice. Experiments on S. Laurence Bay and Beaufort sea ice images show that the proposed method can effectively improve the segmentation accuracy, compared with the region-based MRF algorithm.
Keywords:image of SAR sea ice  image segmentation  Radarsat 1 data  Gibbs sampling
点击此处可从《唐山学院学报》浏览原始摘要信息
点击此处可从《唐山学院学报》下载免费的PDF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