Superpixel guided deep-sparse-representation learning for hyperspectral image classification

This paper presents a new technique for hyperspectral image (HSI) classification by using superpixel guided deep-sparse-representation learning. The proposed technique constructs a hierarchical architecture by exploiting the sparse coding to learn the HSI representation. Specifically, a multiple-lay...

Full description

Saved in:
Bibliographic Details
Main Authors: Fan, Jiayuan, Chen, Tao, Lu, Shijian
Other Authors: School of Computer Science and Engineering
Format: Article
Language:English
Published: 2020
Subjects:
Online Access:https://hdl.handle.net/10356/142926
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-142926
record_format dspace
spelling sg-ntu-dr.10356-1429262020-07-14T01:22:55Z Superpixel guided deep-sparse-representation learning for hyperspectral image classification Fan, Jiayuan Chen, Tao Lu, Shijian School of Computer Science and Engineering Engineering::Computer science and engineering Hyperspectral Classification This paper presents a new technique for hyperspectral image (HSI) classification by using superpixel guided deep-sparse-representation learning. The proposed technique constructs a hierarchical architecture by exploiting the sparse coding to learn the HSI representation. Specifically, a multiple-layer architecture using different superpixel maps is designed, where each superpixel map is generated by downsampling the superpixels gradually along with enlarged spatial regions for labeled samples. In each layer, sparse representation of pixels within every spatial region is computed to construct a histogram via the sum-pooling with $l-{1}$ normalization. Finally, the representations (features) learned from the multiple-layer network are aggregated and trained by a support vector machine classifier. The proposed technique has been evaluated over three public HSI data sets, including the Indian Pines image set, the Salinas image set, and the University of Pavia image set. Experiments show superior performance compared with the state-of-the-art methods. 2020-07-14T01:22:55Z 2020-07-14T01:22:55Z 2017 Journal Article Fan, J., Chen, T., & Lu, S. (2018). Superpixel guided deep-sparse-representation learning for hyperspectral image classification. IEEE Transactions on Circuits and Systems for Video Technology, 28(11), 3163-3173. doi:10.1109/TCSVT.2017.2746684 1051-8215 https://hdl.handle.net/10356/142926 10.1109/TCSVT.2017.2746684 2-s2.0-85028730379 11 28 3163 3173 en IEEE Transactions on Circuits and Systems for Video Technology © 2017 IEEE. All rights reserved.
institution Nanyang Technological University
building NTU Library
country Singapore
collection DR-NTU
language English
topic Engineering::Computer science and engineering
Hyperspectral
Classification
spellingShingle Engineering::Computer science and engineering
Hyperspectral
Classification
Fan, Jiayuan
Chen, Tao
Lu, Shijian
Superpixel guided deep-sparse-representation learning for hyperspectral image classification
description This paper presents a new technique for hyperspectral image (HSI) classification by using superpixel guided deep-sparse-representation learning. The proposed technique constructs a hierarchical architecture by exploiting the sparse coding to learn the HSI representation. Specifically, a multiple-layer architecture using different superpixel maps is designed, where each superpixel map is generated by downsampling the superpixels gradually along with enlarged spatial regions for labeled samples. In each layer, sparse representation of pixels within every spatial region is computed to construct a histogram via the sum-pooling with $l-{1}$ normalization. Finally, the representations (features) learned from the multiple-layer network are aggregated and trained by a support vector machine classifier. The proposed technique has been evaluated over three public HSI data sets, including the Indian Pines image set, the Salinas image set, and the University of Pavia image set. Experiments show superior performance compared with the state-of-the-art methods.
author2 School of Computer Science and Engineering
author_facet School of Computer Science and Engineering
Fan, Jiayuan
Chen, Tao
Lu, Shijian
format Article
author Fan, Jiayuan
Chen, Tao
Lu, Shijian
author_sort Fan, Jiayuan
title Superpixel guided deep-sparse-representation learning for hyperspectral image classification
title_short Superpixel guided deep-sparse-representation learning for hyperspectral image classification
title_full Superpixel guided deep-sparse-representation learning for hyperspectral image classification
title_fullStr Superpixel guided deep-sparse-representation learning for hyperspectral image classification
title_full_unstemmed Superpixel guided deep-sparse-representation learning for hyperspectral image classification
title_sort superpixel guided deep-sparse-representation learning for hyperspectral image classification
publishDate 2020
url https://hdl.handle.net/10356/142926
_version_ 1681058527985008640