Weighted multifeature hyperspectral image classification via kernel joint sparse representation

详细信息    查看全文

• 作者：Erlei Zhang^a ; ^{zhangerlei0123@163.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Xiangrong Zhang^a ; ^b ; ^{xrzhang@mail.xidian.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Licheng Jiao^a ; ^{lchjiao@mail.xidian.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Hongying Liu^a ; ^{hyliu@xidian.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Shuang Wang^a ; ^{shwang@mail.xidian.edu.cn" class="auth_mail" title="E-mail the corresponding author}Author Vitae ; Biao Hou^a ; ^{avcodec@163.com" class="auth_mail" title="E-mail the corresponding author}Author Vitae
• 关键词：Hyperspectral imagery classification ; kernel trick ; joint sparse representation ; feature extraction
• 刊名：Neurocomputing
• 出版年：2016
• 出版时间：20 February 2016
• 年：2016
• 卷：178
• 期：Complete
• 页码：71-86
• 全文大小：6594 K

文摘

The advantage of using multifeature information for classification has been widely recognized. Representation-based methods with multifeature combination learning have only recently attracted increasing attention for hyperspectral classification. However, nonlinearity in data and the computational load of processing multifeature information and contextual information have been two thorny issues. In this paper, we present a fast joint sparse representation model with multifeature combination learning and its kernel extensions for hyperspectral imagery classification. For several complementary features (spectral, shape, and texture), the proposed model simultaneously acquires a representation vector for each type of feature and encourages the representation vectors to share a common sparsity pattern by imposing the joint sparsity ℓ_row,0${\mathbf{\ell }}_{row,0}$-norm regularization. Thus, the cross-feature information can be taken into account. For different features, different weights are assigned since they may not contribute equally to the final decision. Furthermore, kernel joint sparse representation model is presented to handle nonlinearity in the data. Kernel model projects the data into a high-dimensional space to improve the separability, achieving a better performance than the linear version. At the same time, we incorporate contextual neighborhood knowledge into the learned models. Experiments on several real hyperspectral images indicate that the proposed algorithms with much less memory requirements perform significantly faster than state-of-the-art algorithms, while exhibit highly competitive classification accuracy.