基于稀疏自编码器和边缘保持的Wishart马尔科夫随机场的极化SAR图像分类
投稿时间:2017-07-05  修订日期:2017-11-14  点此下载全文
引用本文:张姝茵,侯彪,焦李成,吴倩.基于稀疏自编码器和边缘保持的Wishart马尔科夫随机场的极化SAR图像分类[J].红外与毫米波学报,2018,37(2):177~183].ZHANG Shu-Yin,HOU Biao,JIAO Li-Cheng,WU Qian.PolSAR image classification based on sparse autoencoder and boundary-preserved WMRF[J].J.Infrared Millim.Waves,2018,37(2):177~183.]
摘要点击次数: 342
全文下载次数: 257
作者单位E-mail
张姝茵 西安电子科技大学 智能感知与图像理解教育部重点实验室 shuyin820217@163.com 
侯彪 西安电子科技大学  
焦李成 西安电子科技大学  
吴倩 西安电子科技大学  
基金项目:国家自然科学基金项目(面上项目,重点项目,重大项目)(61671350, 61573267, 61473215,61572383, 61502369),国家重点基础研究发展计划(973计划)(2013CB329402)
中文摘要:针对极化SAR图像训练样本数目较少问题以及极化SAR图像同质区域较多的特性,提出了一种新的两层分类框架,结合了稀疏自编码器和边缘保持的Wishart马尔科夫随机场对极化SAR图像进行分类.该框架包括个步骤,第一个步骤使用稀疏自编码器来获得一个初始分类;第二个步骤使用边缘保持的Wishart马尔科夫随机场对第一层的分类结果进行修正.在应用Wishart马尔科夫随机场的过程中,由稀疏自编码器分类得到的边缘得以保持,并且提出了新的分类错误纠正策略确保分类的准确性.因此,通过稀疏自编码器得到的精确分类边缘可用于不同的区域并且在应用Wishart马尔科夫的过程中得以保持.和其他分类方法相比,该方法得到较高的分类精度,证明了新方法的有效性.
中文关键词:稀疏自编码器  极化SAR图像  Wishart距离  马尔科夫随机场
 
PolSAR image classification based on sparse autoencoder and boundary-preserved WMRF
Abstract:In order to solve problem of the limited training samples and keep consistency in one region, a new two-level classification scheme is proposed, which combines sparse auto-encoder (SAE) and Boundary-preserved Wishart-markov random fields (BWMRF). In the first layer, an SAE classifier is applied to obtain an initial classification and more accurate regional boundaries. In the second layer, Boundary-preserved Wishart-markov random fields have been used to correct the previous classification results. Meanwhile, the boundaries classified by sparse auto-encoder are preserved, and a new error correction strategy is applied to ensure the classification accuracy. Therefore, accurate region boundaries supplied by SAE are explored to divide different regions, and the coherent in each region will be realized during the BWMRF process. Compared with other classification methods, this method obtains higher classification accuracy and proves the validity of the new scheme.
keywords:spare auto-encoder(SAE), polarimetric synthetic aperture radar(PolSAR) images, Wishart distance, Markov random fields(MRF)
查看全文  查看/发表评论  下载PDF阅读器

版权所有:《红外与毫米波学报》编辑部