• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文OA文献 >Classification of Segments in PolSAR Imagery by Minimum Stochastic Distances Between Wishart Distributions
【2h】

Classification of Segments in PolSAR Imagery by Minimum Stochastic Distances Between Wishart Distributions

机译：用最小随机数分类polsaR图像中的分段   Wishart分布之间的距离

代理获取
本网站仅为用户提供外文OA文献查询和代理获取服务，本网站没有原文。下单后我们将采用程序或人工为您竭诚获取高质量的原文，但由于OA文献来源多样且变更频繁，仍可能出现获取不到、文献不完整或与标题不符等情况，如果获取不到我们将提供退款服务。请知悉。
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

A new classifier for Polarimetric SAR (PolSAR) images is proposed andassessed in this paper. Its input consists of segments, and each one isassigned the class which minimizes a stochastic distance. Assuming the complexWishart model, several stochastic distances are obtained from the h-phi familyof divergences, and they are employed to derive hypothesis test statistics thatare also used in the classification process. This article also presents, as anovelty, analytic expressions for the test statistics based on the followingstochastic distances between complex Wishart models: Kullback-Leibler,Bhattacharyya, Hellinger, R\'enyi, and Chi-Square; also, the test statisticbased on the Bhattacharyya distance between multivariate Gaussian distributionsis presented. The classifier performance is evaluated using simulated and realPolSAR data. The simulated data are based on the complex Wishart model, aimingat the analysis of the proposal well controlled data. The real data refer tothe complex L-band image, acquired during the 1994 SIR-C mission. The resultsof the proposed classifier are compared with those obtained by a Wishartper-pixel/contextual classifier, and we show the better performance of theregion-based classification. The influence of the statistical modeling isassessed by comparing the results using the Bhattacharyya distance betweenmultivariate Gaussian distributions for amplitude data. The results withsimulated data indicate that the proposed classification method has a very goodperformance when the data follow the Wishart model. The proposed classifieralso performs better than the per-pixel/contextual classifier and theBhattacharyya Gaussian distance using SIR-C PolSAR data.
机译：提出并评估了一种新的极化SAR图像分类器。它的输入由段组成，每个段都分配有一个类别，该类别最大程度地减少了随机距离。假设复杂的Wishart模型，从h-phi散度族获得了几个随机距离，并将它们用于导出假设检验统计量，这些统计量也用于分类过程中。本文还以复杂的Wishart模型之间的以下随机距离为基础，为测试统计量提供了解析表达式：Kullback-Leibler，Bhattacharyya，Hellinger，R'enyi和Chi-Square。同时，提出了基于多元高斯分布之间的Bhattacharyya距离的检验统计量。使用模拟和realPolSAR数据评估分类器性能。仿真数据基于复杂的Wishart模型，旨在分析建议的受控数据。实际数据是指在1994年SIR-C任务期间获取的复杂L波段图像。将提出的分类器的结果与Wishartper像素/上下文分类器获得的结果进行比较，我们展示了基于区域的分类的更好性能。通过使用振幅数据的多元高斯分布之间的Bhattacharyya距离比较结果来评估统计建模的影响。仿真结果表明，该方法在遵循Wishart模型的情况下具有很好的分类效果。使用SIR-C PolSAR数据，提出的分类器的性能也优于按像素/上下文分类器和巴氏算子高斯距离。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
代理获取

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号