Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification
Joint Authors
Zhao, Cuijie
Zhao, Hongdong
Wang, Guozhen
Chen, Hong
Source
Issue
Vol. 2020, Issue 2020 (31 Dec. 2020), pp.1-17, 17 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2020-08-26
Country of Publication
Egypt
No. of Pages
17
Main Subjects
Abstract EN
At present, the classification of the hyperspectral image (HSI) based on the deep convolutional network has made great progress.
Due to the high dimensionality of spectral features, limited samples of ground truth, and high nonlinearity of hyperspectral data, effective classification of HSI based on deep convolutional neural networks is still difficult.
This paper proposes a novel deep convolutional network structure, namely, a hybrid depth-separable residual network, for HSI classification, called HDSRN.
The HDSRN model organically combines 3D CNN, 2D CNN, multiresidual network ROR, and depth-separable convolutions to extract deeper abstract features.
On the one hand, due to the addition of multiresidual structures and skip connections, this model can alleviate the problem of over fitting, help the backpropagation of gradients, and extract features more fully.
On the other hand, the depth-separable convolutions are used to learn the spatial feature, which reduces the computational cost and alleviates the decline in accuracy.
Extensive experiments on the popular HSI benchmark datasets show that the performance of the proposed network is better than that of the existing prevalent methods.
American Psychological Association (APA)
Zhao, Cuijie& Zhao, Hongdong& Wang, Guozhen& Chen, Hong. 2020. Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification. Complexity،Vol. 2020, no. 2020, pp.1-17.
https://search.emarefa.net/detail/BIM-1141992
Modern Language Association (MLA)
Zhao, Cuijie…[et al.]. Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification. Complexity No. 2020 (2020), pp.1-17.
https://search.emarefa.net/detail/BIM-1141992
American Medical Association (AMA)
Zhao, Cuijie& Zhao, Hongdong& Wang, Guozhen& Chen, Hong. Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification. Complexity. 2020. Vol. 2020, no. 2020, pp.1-17.
https://search.emarefa.net/detail/BIM-1141992
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-1141992