Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation

Ruoning LI; Kangning CUI; Raymond H. CHAN; Robert J. PLEMMONS

doi:10.1109/IGARSS46834.2022.9883961

Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation

Ruoning LI, Kangning CUI, Raymond H. CHAN^*, Robert J. PLEMMONS

^*Corresponding author for this work

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

8 Citations (Scopus)

Abstract

In this work, a novel algorithm called SVM with Shape-adaptive Reconstruction and Smoothed Total Variation (SaR-SVM-STV) is introduced to classify hyperspectral images, which makes full use of spatial and spectral information. The Shape-adaptive Reconstruction (SaR) is introduced to preprocess each pixel based on the Pearson Correlation be-tween pixels in its shape-adaptive (SA) region. Support Vector Machines (SVMs) are trained to estimate the pixel-wise probability maps of each class. Then the Smoothed Total Variation (STV) model is applied to denoise and generate the final classification map. Experiments show that SaR-SVM-STY outperforms the SVM-STV method with a few training labels, demonstrating the significance of reconstructing hy-perspectral images before classification.

Original language	English
Title of host publication	IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings
Publisher	IEEE
Pages	1368-1371
Number of pages	4
ISBN (Electronic)	9781665427920
ISBN (Print)	9781665427937
DOIs	https://doi.org/10.1109/IGARSS46834.2022.9883961
Publication status	Published - 2022
Externally published	Yes
Event	2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022 - Kuala Lumpur, Malaysia Duration: 17 Jul 2022 → 22 Jul 2022

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Volume	2022-July

Conference

Conference	2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022
Country/Territory	Malaysia
City	Kuala Lumpur
Period	17/07/22 → 22/07/22

Bibliographical note

Publisher Copyright:
© 2022 IEEE.

Funding

This work was supported in part by HKRGC Grants No. CUHK14301718, CityU11301120, C1013-21GF, CityU Grant 9380101.

Keywords

Hyperspectral Image Classification
Image Reconstruction
Semi-supervised Learning
Smoothed To-tal Variation
Support Vector Machines

Access to Document

10.1109/IGARSS46834.2022.9883961

Cite this

LI, R., CUI, K., CHAN, R. H., & PLEMMONS, R. J. (2022). Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation. In IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings (pp. 1368-1371). (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2022-July). IEEE. https://doi.org/10.1109/IGARSS46834.2022.9883961

@inproceedings{5fbddcfbaa484337bb02d898a884db2f,

title = "Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation",

abstract = "In this work, a novel algorithm called SVM with Shape-adaptive Reconstruction and Smoothed Total Variation (SaR-SVM-STV) is introduced to classify hyperspectral images, which makes full use of spatial and spectral information. The Shape-adaptive Reconstruction (SaR) is introduced to preprocess each pixel based on the Pearson Correlation be-tween pixels in its shape-adaptive (SA) region. Support Vector Machines (SVMs) are trained to estimate the pixel-wise probability maps of each class. Then the Smoothed Total Variation (STV) model is applied to denoise and generate the final classification map. Experiments show that SaR-SVM-STY outperforms the SVM-STV method with a few training labels, demonstrating the significance of reconstructing hy-perspectral images before classification.",

keywords = "Hyperspectral Image Classification, Image Reconstruction, Semi-supervised Learning, Smoothed To-tal Variation, Support Vector Machines",

author = "Ruoning LI and Kangning CUI and CHAN, \{Raymond H.\} and PLEMMONS, \{Robert J.\}",

note = "Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022 ; Conference date: 17-07-2022 Through 22-07-2022",

year = "2022",

doi = "10.1109/IGARSS46834.2022.9883961",

language = "English",

isbn = "9781665427937",

series = "International Geoscience and Remote Sensing Symposium (IGARSS)",

publisher = "IEEE",

pages = "1368--1371",

booktitle = "IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings",

}

LI, R, CUI, K, CHAN, RH & PLEMMONS, RJ 2022, Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation. in IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. International Geoscience and Remote Sensing Symposium (IGARSS), vol. 2022-July, IEEE, pp. 1368-1371, 2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022, Kuala Lumpur, Malaysia, 17/07/22. https://doi.org/10.1109/IGARSS46834.2022.9883961

Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation. / LI, Ruoning; CUI, Kangning; CHAN, Raymond H. et al.
IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. IEEE, 2022. p. 1368-1371 (International Geoscience and Remote Sensing Symposium (IGARSS); Vol. 2022-July).

Research output: Book Chapters | Papers in Conference Proceedings › Conference paper (refereed) › Research › peer-review

TY - GEN

T1 - Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation

AU - LI, Ruoning

AU - CUI, Kangning

AU - CHAN, Raymond H.

AU - PLEMMONS, Robert J.

PY - 2022

Y1 - 2022

N2 - In this work, a novel algorithm called SVM with Shape-adaptive Reconstruction and Smoothed Total Variation (SaR-SVM-STV) is introduced to classify hyperspectral images, which makes full use of spatial and spectral information. The Shape-adaptive Reconstruction (SaR) is introduced to preprocess each pixel based on the Pearson Correlation be-tween pixels in its shape-adaptive (SA) region. Support Vector Machines (SVMs) are trained to estimate the pixel-wise probability maps of each class. Then the Smoothed Total Variation (STV) model is applied to denoise and generate the final classification map. Experiments show that SaR-SVM-STY outperforms the SVM-STV method with a few training labels, demonstrating the significance of reconstructing hy-perspectral images before classification.

AB - In this work, a novel algorithm called SVM with Shape-adaptive Reconstruction and Smoothed Total Variation (SaR-SVM-STV) is introduced to classify hyperspectral images, which makes full use of spatial and spectral information. The Shape-adaptive Reconstruction (SaR) is introduced to preprocess each pixel based on the Pearson Correlation be-tween pixels in its shape-adaptive (SA) region. Support Vector Machines (SVMs) are trained to estimate the pixel-wise probability maps of each class. Then the Smoothed Total Variation (STV) model is applied to denoise and generate the final classification map. Experiments show that SaR-SVM-STY outperforms the SVM-STV method with a few training labels, demonstrating the significance of reconstructing hy-perspectral images before classification.

KW - Hyperspectral Image Classification

KW - Image Reconstruction

KW - Semi-supervised Learning

KW - Smoothed To-tal Variation

KW - Support Vector Machines

UR - https://www.scopus.com/pages/publications/85140368189

U2 - 10.1109/IGARSS46834.2022.9883961

DO - 10.1109/IGARSS46834.2022.9883961

M3 - Conference paper (refereed)

AN - SCOPUS:85140368189

SN - 9781665427937

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 1368

EP - 1371

BT - IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings

PB - IEEE

T2 - 2022 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2022

Y2 - 17 July 2022 through 22 July 2022

ER -

LI R, CUI K, CHAN RH, PLEMMONS RJ. Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation. In IGARSS 2022: 2022 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. IEEE. 2022. p. 1368-1371. (International Geoscience and Remote Sensing Symposium (IGARSS)). doi: 10.1109/IGARSS46834.2022.9883961

Classification of Hyperspectral Images Using SVM with Shape-Adaptive Reconstruction and Smoothed Total Variation

Abstract

Publication series

Conference

Bibliographical note

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this