unsupervised classification of multifrequency and.pptx

下载文档

3
0
约 16页
2017-04-11 发布于江苏
举报
版权申诉
保障服务

unsupervised classification of multifrequency and.pptx

1、本文档共16页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。

unsupervised classification of multifrequency and

SAR Image Classification 2017-4-11 1 classification of multi-frequency classification Name: unsupervised classification of multi-frequency and fully polarimetric SAR images based on the H/A/Alpha-Wishart classifier Author: Laurent Ferro-Famil, Eric Pottier,2001 2017-4-11 2 classification of multi-frequency Single-frequency polarimetric SAR data statistics 2017-4-11 3 classification of multi-frequency Dual-frequency polarimetric SAR data statistics A. dual polarimetric representation 2017-4-11 4 classification of multi-frequency B. cross-correlation Matrix statistics 2017-4-11 5 classification of multi-frequency Single image classification procedure A. classification algorithm B. the H-alpha classification scheme 2017-4-11 6 classification of multi-frequency C. Application to PolSAR data Single frequency processing P band and L band obtained with the unsupervised Wishart classifier 2017-4-11 7 classification of multi-frequency Multi-frequency classification procedure A. Dual frequency image ML classification 1)classification algorithm 2017-4-11 8 classification of multi-frequency Classification procedures 2017-4-11 9 classification of multi-frequency 2)reduction of the number of classes 步骤1）将图像分成64类，步骤2）把64类聚类成想要的类别数，例如16类。 2017-4-11 10 classification of multi-frequency 3) Application to PolSAR data With the number of classes reduced to 16 The class number reduced technique permits to merge classes of close characteristics and produces an accurate distribution of the differently aged trees in the forest It is important to note that no auuumptions were mage concerning the level of cross-correlation between the different data sets, the classification taking into account this information during the entire process. 2017-4-11 11 classification of multi-frequency B. Dual frequency image classification using the cross-correlation information 1)classification algorithm 2017-4-11 12 classification of multi-frequency Decision rules 2017-4-11 13 classification of multi-frequenc