FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer

Remote sensing scene classification (RSSC) is to accurately assign semantic labels to remote sensing images by analyzing scene contents. Recently, many algorithms have made significant progress in improving the classification accuracy of RSSC. However, a large number of parameters and floating point...

Full description

Saved in:
Bibliographic Details
Main Authors: Wei Wu, Xianbin Hu, Zhu Li, Xueliang Luo
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11044321/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849744408972361728
author Wei Wu
Xianbin Hu
Zhu Li
Xueliang Luo
author_facet Wei Wu
Xianbin Hu
Zhu Li
Xueliang Luo
author_sort Wei Wu
collection DOAJ
description Remote sensing scene classification (RSSC) is to accurately assign semantic labels to remote sensing images by analyzing scene contents. Recently, many algorithms have made significant progress in improving the classification accuracy of RSSC. However, a large number of parameters and floating point operations are needed to achieve that end in those approaches, resulting in high complexity. To address the issue, we propose a novel RSSC algorithm, dubbed FishermaskFormer, which aggressively decimates features in the convolutional backbone via a novel masking operation with a proposed fisher discriminant analysis criterion, and then designs a lightweight transformer block to drive the classification loss. This is aimed at offering a flexible and effective framework for preserving classification accuracy while significantly reducing the complexity. The proposed transformer design employs a new grouping index that assigns multiheaded transformer groups by maximizing the information interactions in each group. Compared with leading lightweight RSSC methods, experimental results show this proposed framework achieves higher classification accuracy while having the similar low complexity.
format Article
id doaj-art-5cc5b9b1f8f14e19a31a43231587f03e
institution DOAJ
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-5cc5b9b1f8f14e19a31a43231587f03e2025-08-20T03:17:24ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118158291584410.1109/JSTARS.2025.358137811044321FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked TransformerWei Wu0https://orcid.org/0000-0002-1301-2575Xianbin Hu1Zhu Li2https://orcid.org/0000-0002-8246-177XXueliang Luo3State Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, ChinaState Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, ChinaDepartment of Computer Science & Electrical Engineering, University of Missouri, Kansas City, MO, USAState Key Laboratory of Integrated Services Networks, Xidian University, Xi’an, ChinaRemote sensing scene classification (RSSC) is to accurately assign semantic labels to remote sensing images by analyzing scene contents. Recently, many algorithms have made significant progress in improving the classification accuracy of RSSC. However, a large number of parameters and floating point operations are needed to achieve that end in those approaches, resulting in high complexity. To address the issue, we propose a novel RSSC algorithm, dubbed FishermaskFormer, which aggressively decimates features in the convolutional backbone via a novel masking operation with a proposed fisher discriminant analysis criterion, and then designs a lightweight transformer block to drive the classification loss. This is aimed at offering a flexible and effective framework for preserving classification accuracy while significantly reducing the complexity. The proposed transformer design employs a new grouping index that assigns multiheaded transformer groups by maximizing the information interactions in each group. Compared with leading lightweight RSSC methods, experimental results show this proposed framework achieves higher classification accuracy while having the similar low complexity.https://ieeexplore.ieee.org/document/11044321/Classification accuracyfisher discriminant analysis criterionmodel complexityremote sensing scene classification (RSSC)transformer
spellingShingle Wei Wu
Xianbin Hu
Zhu Li
Xueliang Luo
FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Classification accuracy
fisher discriminant analysis criterion
model complexity
remote sensing scene classification (RSSC)
transformer
title FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer
title_full FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer
title_fullStr FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer
title_full_unstemmed FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer
title_short FishermaskFormer: Lightweight Remote Sensing Scene Classification With Masked Transformer
title_sort fishermaskformer lightweight remote sensing scene classification with masked transformer
topic Classification accuracy
fisher discriminant analysis criterion
model complexity
remote sensing scene classification (RSSC)
transformer
url https://ieeexplore.ieee.org/document/11044321/
work_keys_str_mv AT weiwu fishermaskformerlightweightremotesensingsceneclassificationwithmaskedtransformer
AT xianbinhu fishermaskformerlightweightremotesensingsceneclassificationwithmaskedtransformer
AT zhuli fishermaskformerlightweightremotesensingsceneclassificationwithmaskedtransformer
AT xueliangluo fishermaskformerlightweightremotesensingsceneclassificationwithmaskedtransformer