首页    期刊浏览 2022年08月08日 星期一
登录注册

文章基本信息

  • 标题:The Study on Linear Mixed Parcel Unmixing for Classification
  • 本地全文:下载
  • 作者:J. Zhang ; Y. Pan ; L. Li
  • 期刊名称:ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
  • 印刷版ISSN:2194-9042
  • 电子版ISSN:2194-9050
  • 出版年度:2008
  • 卷号:XXXVII Part B7
  • 页码:621-624
  • 出版社:Copernicus Publications
  • 摘要:Aiming at the disadvantage of hard per-parcel classification which can't solve the difficulty of mixed parcel resulting in the low accuracy, a new method of soft per-parcel classification is presented, that is linear mixed parcel unmixing. Based on the linear spectral theory for the parcel unmixing, the predicted fraction value is assigned to a parcel. The RMSE results show that the accuracy of this method is similar to the traditional linear spectral unmxing method, and is higher than that of hard per-parcel classification. There are two advantages about this method. Firstly, this method incorporating more than one pixel information ensures the result stability. Secondly, the problem of mixed parcel, which disturbs the hard per-parcel classification, can be solved. All of above improve the per-parcel classification accuracy
  • 关键词:Linear mixed parcel unmxing; Linear spectral unmixing; Per-parcel classification; ISODATA
国家哲学社会科学文献中心版权所有