Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence

D. E. Turan; E. Aptoula; A. Erturk; G. Taskin

doi:10.1109/IGARSS52108.2023.10282341

Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence

D. E. Turan^*
, E. Aptoula
, A. Erturk
, G. Taskin

^*Corresponding author for this work

Disaster Management Institute

Sabanci University
Kocaeli University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

7 Citations (Scopus)

Abstract

This study addresses the explainability challenges of deep-learning models in the context of hyperspectral remote sensing image classification. Three prominent explainable artificial intelligence methods, namely GradCAM, GradCAM++, and Guided Backpropagation, have been employed in order to comprehend the decision-making process of a typical convolutional neural network model during spatial-spectral hyperspectral image classification. The experiments that have been conducted investigate the impact of pixel patch sizes on spatial attention, as well as spectral band importance. The findings provide insights into the behavior of both convolutional neural networks, as well as the comparative performance of explainability techniques.

Original language	English
Title of host publication	IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	5950-5953
Number of pages	4
ISBN (Electronic)	9798350320107
DOIs	https://doi.org/10.1109/IGARSS52108.2023.10282341
Publication status	Published - 2023
Event	2023 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023 - Pasadena, United States Duration: 16 Jul 2023 → 21 Jul 2023

Publication series

Name	International Geoscience and Remote Sensing Symposium (IGARSS)
Volume	2023-July
ISSN (Electronic)	2153-6996

Conference

Conference	2023 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023
Country/Territory	United States
City	Pasadena
Period	16/07/23 → 21/07/23

Bibliographical note

Publisher Copyright:
© 2023 IEEE.

Keywords

Explainable artificial intelligence
GradCam
guided backpropagation
hyperspectral images
interpretability

Access to Document

10.1109/IGARSS52108.2023.10282341

Cite this

Turan, D. E., Aptoula, E., Erturk, A., & Taskin, G. (2023). Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence. In IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings (pp. 5950-5953). (International Geoscience and Remote Sensing Symposium (IGARSS) ; Vol. 2023-July). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IGARSS52108.2023.10282341

Turan, D. E. ; Aptoula, E. ; Erturk, A. et al. / Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence. IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 5950-5953 (International Geoscience and Remote Sensing Symposium (IGARSS) ).

@inproceedings{42749acdb71147f8833560dbe0957f33,

title = "Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence",

abstract = "This study addresses the explainability challenges of deep-learning models in the context of hyperspectral remote sensing image classification. Three prominent explainable artificial intelligence methods, namely GradCAM, GradCAM++, and Guided Backpropagation, have been employed in order to comprehend the decision-making process of a typical convolutional neural network model during spatial-spectral hyperspectral image classification. The experiments that have been conducted investigate the impact of pixel patch sizes on spatial attention, as well as spectral band importance. The findings provide insights into the behavior of both convolutional neural networks, as well as the comparative performance of explainability techniques.",

keywords = "Explainable artificial intelligence, GradCam, guided backpropagation, hyperspectral images, interpretability",

author = "Turan, \{D. E.\} and E. Aptoula and A. Erturk and G. Taskin",

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023 ; Conference date: 16-07-2023 Through 21-07-2023",

year = "2023",

doi = "10.1109/IGARSS52108.2023.10282341",

language = "English",

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

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "5950--5953",

booktitle = "IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings",

address = "United States",

}

Turan, DE, Aptoula, E, Erturk, A & Taskin, G 2023, Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence. in IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. International Geoscience and Remote Sensing Symposium (IGARSS) , vol. 2023-July, Institute of Electrical and Electronics Engineers Inc., pp. 5950-5953, 2023 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023, Pasadena, United States, 16/07/23. https://doi.org/10.1109/IGARSS52108.2023.10282341

Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence. / Turan, D. E.; Aptoula, E.; Erturk, A. et al.
IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. Institute of Electrical and Electronics Engineers Inc., 2023. p. 5950-5953 (International Geoscience and Remote Sensing Symposium (IGARSS) ; Vol. 2023-July).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence

AU - Turan, D. E.

AU - Aptoula, E.

AU - Erturk, A.

AU - Taskin, G.

PY - 2023

Y1 - 2023

N2 - This study addresses the explainability challenges of deep-learning models in the context of hyperspectral remote sensing image classification. Three prominent explainable artificial intelligence methods, namely GradCAM, GradCAM++, and Guided Backpropagation, have been employed in order to comprehend the decision-making process of a typical convolutional neural network model during spatial-spectral hyperspectral image classification. The experiments that have been conducted investigate the impact of pixel patch sizes on spatial attention, as well as spectral band importance. The findings provide insights into the behavior of both convolutional neural networks, as well as the comparative performance of explainability techniques.

AB - This study addresses the explainability challenges of deep-learning models in the context of hyperspectral remote sensing image classification. Three prominent explainable artificial intelligence methods, namely GradCAM, GradCAM++, and Guided Backpropagation, have been employed in order to comprehend the decision-making process of a typical convolutional neural network model during spatial-spectral hyperspectral image classification. The experiments that have been conducted investigate the impact of pixel patch sizes on spatial attention, as well as spectral band importance. The findings provide insights into the behavior of both convolutional neural networks, as well as the comparative performance of explainability techniques.

KW - Explainable artificial intelligence

KW - GradCam

KW - guided backpropagation

KW - hyperspectral images

KW - interpretability

UR - https://www.scopus.com/pages/publications/85178383641

U2 - 10.1109/IGARSS52108.2023.10282341

DO - 10.1109/IGARSS52108.2023.10282341

M3 - Conference contribution

AN - SCOPUS:85178383641

T3 - International Geoscience and Remote Sensing Symposium (IGARSS)

SP - 5950

EP - 5953

BT - IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2023 IEEE International Geoscience and Remote Sensing Symposium, IGARSS 2023

Y2 - 16 July 2023 through 21 July 2023

ER -

Turan DE, Aptoula E, Erturk A, Taskin G. Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence. In IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Proceedings. Institute of Electrical and Electronics Engineers Inc. 2023. p. 5950-5953. (International Geoscience and Remote Sensing Symposium (IGARSS) ). doi: 10.1109/IGARSS52108.2023.10282341

Interpreting Hyperspectral Remote Sensing Image Classification Methods Via Explainable Artificial Intelligence

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this