High-resolution remote sensing image change detection based on Fourier feature interaction and multi-scale perception.

Chen, Yongqi; Feng, Shou; Zhao, Chunhui; Su, Nan; Li, Wei; Tao, Ran; Ren, Jinchang

Abstract

As a significant means of Earth observation, change detection in high-resolution remote sensing images has received extensive attention. Nevertheless, the variability in imaging conditions introduces style discrepancies and a range of pseudo change regions between bi-temporal image pairs. Furthermore, changing objects possess diverse morphological representations, which makes accurately identifying change areas and delineating their boundaries within complex object distributions increasingly difficult. In response to the aforementioned challenges, we propose Fourier feature interaction and multi-scale perception (FIMP) model for effective change detection. To mitigate the impact of style discrepancies, FIMP employs the Fourier transform to adaptively filter bi-temporal features in the frequency domain, whilst mining the optimized bi-temporal features relevant to the change detection task. To enhance the ability to recognize multi-scale changing objects, FIMP aggregates and emphasizes the change areas with the introduced temporal change enhancement module (TCEM). By utilizing the U-fusion change perception module (UCPM) to perform multi-level bidirectional fusion of change features at different scales, FIMP can further enhance the ability to delineate complex semantic change boundaries. Experiments on three public datasets shows that our approach outperforms seven state-of-the-art methods.

Citation

CHEN, Y., FENG, S., ZHAO, C., SU, N., LI, W., TAO, R. and REN, J. 2024. High-resolution remote sensing image change detection based on Fourier feature interaction and multi-scale perception. IEEE transactions on geoscience and remote sensing [online], 62, article number 3500073. Available from: https://doi.org/10.1109/TGRS.2024.3500073

• Journal Article Type: Article
• Acceptance Date: Nov 18, 2024
• Online Publication Date: Nov 18, 2024
• Publication Date: Dec 31, 2024
• Deposit Date: Nov 21, 2024
• Publicly Available Date: Nov 21, 2024
• Journal: IEEE transactions on geoscience and remote sensing
• Print ISSN: 0196-2892
• Electronic ISSN: 1558-0644
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Volume: 62
• Article Number: 5539115
• DOI: https://doi.org/10.1109/TGRS.2024.3500073
• Keywords: Change detection; High-resolution remote sensing image; Fourier feature interaction; Multi-scale change feature
• Public URL: https://rgu-repository.worktribe.com/output/2584576