Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images

Semantic Change Detection (SCD) in Remote Sensing Images (RSI) aims to identify changes in the type of Land Cover/Land Use (LCLU). The “from-to” information of the acquired image has more profound practical significance than Binary Change Detection (BCD). However, most deep lea...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuhang Zhang, Wuxia Zhang, Songtao Ding, Siyuan Wu, Xiaoqiang Lu
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/10979855/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849732096826802176
author Yuhang Zhang
Wuxia Zhang
Songtao Ding
Siyuan Wu
Xiaoqiang Lu
author_facet Yuhang Zhang
Wuxia Zhang
Songtao Ding
Siyuan Wu
Xiaoqiang Lu
author_sort Yuhang Zhang
collection DOAJ
description Semantic Change Detection (SCD) in Remote Sensing Images (RSI) aims to identify changes in the type of Land Cover/Land Use (LCLU). The “from-to” information of the acquired image has more profound practical significance than Binary Change Detection (BCD). However, most deep learning-based SCD algorithms do not fully exploit the spatial-temporal information of multilevel features, leading to challenges in extracting LCLU features in complex scenes. To address these issues, we propose a Spatial-Temporal Semantic Feature Interaction Network (STS-FINet) to improve the performance of SCD in RSI. The proposed STS-FINet comprises a Multi-Scale Feature Extraction Encoder (MS-FEE), a Transformer-based Multilevel Feature Interaction module (TML-FI), and a Multilevel Feature Fusion Decoder (ML-FFD). The MS-FEE extracts deep semantic and differential information from the RSI. The TML-FI is designed to mine the spatial-temporal information by extracting long-range dependencies and spatial information from multilevel features to improve spatial perception. Moreover, Mixed Spatial Reasoning Convolution block (MixSrc) is presented to enrich the spatial information by extracting the multiscale features, thus improving the model's capability to interpret complex scenes. Finally, ML-FFD integrates the multilevel features, resulting in the generation of the semantic change map. The effectiveness of the proposed STS-FINet is verified on two high-resolution RSI datasets. Experimental results show that the proposed STS-FINet achieves better change detection performance than SOTA methods.
format Article
id doaj-art-a81dc3a33aa04e2f82c8144fd4fab0bc
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-a81dc3a33aa04e2f82c8144fd4fab0bc2025-08-20T03:08:20ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118120901210210.1109/JSTARS.2025.356538310979855Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing ImagesYuhang Zhang0Wuxia Zhang1https://orcid.org/0000-0002-0759-2489Songtao Ding2Siyuan Wu3Xiaoqiang Lu4Shaanxi Key Laboratory of Network Data Analysis and Intelligent Processing, School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Xi'an, ChinaShaanxi Key Laboratory of Network Data Analysis and Intelligent Processing, School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Xi'an, ChinaShaanxi Key Laboratory of Network Data Analysis and Intelligent Processing, School of Computer Science and Technology, Xi'an University of Posts and Telecommunications, Xi'an, ChinaCollege of Computer Science and Engineering, Xi'an University of Technology, Xi'an, Shaanxi, ChinaCollege of Physics and Information Engineering, Fuzhou University, Fuzhou, ChinaSemantic Change Detection (SCD) in Remote Sensing Images (RSI) aims to identify changes in the type of Land Cover/Land Use (LCLU). The “from-to” information of the acquired image has more profound practical significance than Binary Change Detection (BCD). However, most deep learning-based SCD algorithms do not fully exploit the spatial-temporal information of multilevel features, leading to challenges in extracting LCLU features in complex scenes. To address these issues, we propose a Spatial-Temporal Semantic Feature Interaction Network (STS-FINet) to improve the performance of SCD in RSI. The proposed STS-FINet comprises a Multi-Scale Feature Extraction Encoder (MS-FEE), a Transformer-based Multilevel Feature Interaction module (TML-FI), and a Multilevel Feature Fusion Decoder (ML-FFD). The MS-FEE extracts deep semantic and differential information from the RSI. The TML-FI is designed to mine the spatial-temporal information by extracting long-range dependencies and spatial information from multilevel features to improve spatial perception. Moreover, Mixed Spatial Reasoning Convolution block (MixSrc) is presented to enrich the spatial information by extracting the multiscale features, thus improving the model's capability to interpret complex scenes. Finally, ML-FFD integrates the multilevel features, resulting in the generation of the semantic change map. The effectiveness of the proposed STS-FINet is verified on two high-resolution RSI datasets. Experimental results show that the proposed STS-FINet achieves better change detection performance than SOTA methods.https://ieeexplore.ieee.org/document/10979855/Multiscale analysisremote sensingsemantic change detection (SCD)transformer
spellingShingle Yuhang Zhang
Wuxia Zhang
Songtao Ding
Siyuan Wu
Xiaoqiang Lu
Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Multiscale analysis
remote sensing
semantic change detection (SCD)
transformer
title Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images
title_full Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images
title_fullStr Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images
title_full_unstemmed Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images
title_short Spatial-Temporal Semantic Feature Interaction Network for Semantic Change Detection in Remote Sensing Images
title_sort spatial temporal semantic feature interaction network for semantic change detection in remote sensing images
topic Multiscale analysis
remote sensing
semantic change detection (SCD)
transformer
url https://ieeexplore.ieee.org/document/10979855/
work_keys_str_mv AT yuhangzhang spatialtemporalsemanticfeatureinteractionnetworkforsemanticchangedetectioninremotesensingimages
AT wuxiazhang spatialtemporalsemanticfeatureinteractionnetworkforsemanticchangedetectioninremotesensingimages
AT songtaoding spatialtemporalsemanticfeatureinteractionnetworkforsemanticchangedetectioninremotesensingimages
AT siyuanwu spatialtemporalsemanticfeatureinteractionnetworkforsemanticchangedetectioninremotesensingimages
AT xiaoqianglu spatialtemporalsemanticfeatureinteractionnetworkforsemanticchangedetectioninremotesensingimages