在线客服: 点击这里给我发消息  新用户使用步骤:会员注册→充值→重新登入→进入资源
标题:Semi-Supervised Deep Learning Classification for Hyperspectral Image Based on Dual-Strategy Sample Selection
时间:2019-01-11 22:08:51
DOI:10.3390/rs10040574
作者:Fang, Bei;Li, Ying;Zhang, Haokui;Chan, Jonathan
出版源: 《Remote Sensing》 ,2018 ,10 (4) :574-
摘要:Semi-Supervised Deep Learning Classification for Hyperspectral Image Based on Dual-Strategy Sample SelectionThis paper studies the classification problem of hyperspectral image (HSI). Inspired by the great success of deep neural networks in ...
大小:28020 kb
页数:24 PAGES
下载: 点击下载
预览:

浏览器不支持嵌入PDF阅读,打开新页面在线阅读

目录:
  • Introduction
  • Method
    • Overview
    • Networks Architectures Based on Spectral and Spatial Features
    • Dual-Strategy Sample Selection Co-Training
      • New Sample Selection Mechanism Based on Spectral Feature
      • Sample Selection Mechanism Based on Spatial Feature
      • Co-Training
  • Experimental Results and Analyses
    • Dataset Description and Experimental Settings
    • Experimental Results on the AVIRIS Indian Pines Dataset
    • Experimental Results on the ROSIS-03 University of Pavia Dataset
    • Experimental Results on the AVIRIS Salinas Valley Dataset
    • Experimental Results on Hyperion Dataset
  • Discussion
    • Influence of Network Hyper-Parameters
    • Effect of the Number of Iterations in Co-Training
    • Sample Selection Mechanism Analysis in Co-Training
  • Conclusions
  • References

本页内容由网络收集而来,版权归原创者所有,如有侵权请及时联系