Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network

Yi Miao; Jun Jie Wang; Meng Meng Zhang; Xiao Ming Xie; Wei Li

doi:10.1007/978-981-97-9919-0_1

Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network

Yi Miao^*, Jun Jie Wang^*, Meng Meng Zhang, Xiao Ming Xie^*, Wei Li

^*Corresponding author for this work

School of Information and Electronics

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

Abstract

Classifying optical remote sensing images is a crucial subtask in remote sensing image interpretation. With the advancement of remote sensing technology, more complex scene data can be acquired. However, remote sensing image classification faces challenges due to inter-class image similarities and intra-class diversities, as well as scene significant scale differences in these scenes. To address these issues, this paper proposes a novel method for remote sensing scene classification using a Multi-scale Local Attention Network (MSLANet). The method integrates dual attention mechanisms of channel and space in feature extraction to enhance model sensitivity to regions of interest. Multi-scale features are also fused, this is an improvement that aimed at better fusing image features to enhance model robustness. Our model achieves state-of-the-art performance on three publicly available scene classification datasets.

Original language	English
Title of host publication	Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers
Editors	Yongtian Wang, Hua Huang
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	1-15
Number of pages	15
ISBN (Print)	9789819799183
DOIs	https://doi.org/10.1007/978-981-97-9919-0_1
Publication status	Published - 2025
Event	19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024 - Beijing, China Duration: 16 Aug 2024 → 18 Aug 2024

Publication series

Name	Communications in Computer and Information Science
Volume	2302 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024
Country/Territory	China
City	Beijing
Period	16/08/24 → 18/08/24

Keywords

Attention Mechanism
Multi-scale local attention
Remote sensing
Scene classification

Access to Document

10.1007/978-981-97-9919-0_1

Cite this

Miao, Y., Wang, J. J., Zhang, M. M., Xie, X. M., & Li, W. (2025). Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network. In Y. Wang, & H. Huang (Eds.), Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers (pp. 1-15). (Communications in Computer and Information Science; Vol. 2302 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-97-9919-0_1

Miao, Yi ; Wang, Jun Jie ; Zhang, Meng Meng et al. / Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network. Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers. editor / Yongtian Wang ; Hua Huang. Springer Science and Business Media Deutschland GmbH, 2025. pp. 1-15 (Communications in Computer and Information Science).

@inproceedings{f84aaf4f32914a248d99a30ebefcfe55,

title = "Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network",

abstract = "Classifying optical remote sensing images is a crucial subtask in remote sensing image interpretation. With the advancement of remote sensing technology, more complex scene data can be acquired. However, remote sensing image classification faces challenges due to inter-class image similarities and intra-class diversities, as well as scene significant scale differences in these scenes. To address these issues, this paper proposes a novel method for remote sensing scene classification using a Multi-scale Local Attention Network (MSLANet). The method integrates dual attention mechanisms of channel and space in feature extraction to enhance model sensitivity to regions of interest. Multi-scale features are also fused, this is an improvement that aimed at better fusing image features to enhance model robustness. Our model achieves state-of-the-art performance on three publicly available scene classification datasets.",

keywords = "Attention Mechanism, Multi-scale local attention, Remote sensing, Scene classification",

author = "Yi Miao and Wang, {Jun Jie} and Zhang, {Meng Meng} and Xie, {Xiao Ming} and Wei Li",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; 19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024 ; Conference date: 16-08-2024 Through 18-08-2024",

year = "2025",

doi = "10.1007/978-981-97-9919-0_1",

language = "English",

isbn = "9789819799183",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "1--15",

editor = "Yongtian Wang and Hua Huang",

booktitle = "Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers",

address = "Germany",

}

Miao, Y, Wang, JJ, Zhang, MM, Xie, XM & Li, W 2025, Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network. in Y Wang & H Huang (eds), Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers. Communications in Computer and Information Science, vol. 2302 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 1-15, 19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024, Beijing, China, 16/08/24. https://doi.org/10.1007/978-981-97-9919-0_1

Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network. / Miao, Yi; Wang, Jun Jie; Zhang, Meng Meng et al.
Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers. ed. / Yongtian Wang; Hua Huang. Springer Science and Business Media Deutschland GmbH, 2025. p. 1-15 (Communications in Computer and Information Science; Vol. 2302 CCIS).

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

TY - GEN

T1 - Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network

AU - Miao, Yi

AU - Wang, Jun Jie

AU - Zhang, Meng Meng

AU - Xie, Xiao Ming

AU - Li, Wei

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

PY - 2025

Y1 - 2025

N2 - Classifying optical remote sensing images is a crucial subtask in remote sensing image interpretation. With the advancement of remote sensing technology, more complex scene data can be acquired. However, remote sensing image classification faces challenges due to inter-class image similarities and intra-class diversities, as well as scene significant scale differences in these scenes. To address these issues, this paper proposes a novel method for remote sensing scene classification using a Multi-scale Local Attention Network (MSLANet). The method integrates dual attention mechanisms of channel and space in feature extraction to enhance model sensitivity to regions of interest. Multi-scale features are also fused, this is an improvement that aimed at better fusing image features to enhance model robustness. Our model achieves state-of-the-art performance on three publicly available scene classification datasets.

AB - Classifying optical remote sensing images is a crucial subtask in remote sensing image interpretation. With the advancement of remote sensing technology, more complex scene data can be acquired. However, remote sensing image classification faces challenges due to inter-class image similarities and intra-class diversities, as well as scene significant scale differences in these scenes. To address these issues, this paper proposes a novel method for remote sensing scene classification using a Multi-scale Local Attention Network (MSLANet). The method integrates dual attention mechanisms of channel and space in feature extraction to enhance model sensitivity to regions of interest. Multi-scale features are also fused, this is an improvement that aimed at better fusing image features to enhance model robustness. Our model achieves state-of-the-art performance on three publicly available scene classification datasets.

KW - Attention Mechanism

KW - Multi-scale local attention

KW - Remote sensing

KW - Scene classification

UR - http://www.scopus.com/inward/record.url?scp=85214230061&partnerID=8YFLogxK

U2 - 10.1007/978-981-97-9919-0_1

DO - 10.1007/978-981-97-9919-0_1

M3 - Conference contribution

AN - SCOPUS:85214230061

SN - 9789819799183

T3 - Communications in Computer and Information Science

SP - 1

EP - 15

BT - Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers

A2 - Wang, Yongtian

A2 - Huang, Hua

PB - Springer Science and Business Media Deutschland GmbH

T2 - 19th Chinese Conference on Image and Graphics Technologies and Applications, IGTA 2024

Y2 - 16 August 2024 through 18 August 2024

ER -

Miao Y, Wang JJ, Zhang MM, Xie XM, Li W. Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network. In Wang Y, Huang H, editors, Image and Graphics Technologies and Applications - 19th Chinese Conference, IGTA 2024, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2025. p. 1-15. (Communications in Computer and Information Science). doi: 10.1007/978-981-97-9919-0_1

Remote Sensing Scene Classification Method Based on Multi-scale Local Attention Network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this