Hyperspectral image reconstruction by deep convolutional neural network for classification

详细信息    查看全文

• 作者：Yunsong Li^a ; ^bAuthor Vitae ; Weiying Xie^a ; ^b ; ^{zwpxwy@126.com}Author Vitae ; Huaqing Li^a ; ^bAuthor Vitae
• 关键词：Hyperspectral imagery ; Deep convolutional neural network ; Extreme learning machine ; Reconstruction ; Band selection ; Pattern classification
• 刊名：Pattern Recognition
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：63
• 期：Complete
• 页码：371-383
• 全文大小：6239 K
• 卷排序：63

文摘

The feature enhancement step is introduced in the proposed HRCNN model because that the HSI is contaminated with noise. A new band selection method based on spatial features is proposed in this training model. A new exploration to combine CNN with ELM for HSI classification under a small training set is achieved.