Tongfei Liu

Heterogeneous remote sensing change detection (CD) is a research hotspot in the field of remote sensing. However, due to the differences between different modalities, heterogeneous remote sensing images (HRSI) are often difficult to compare and analyze directly. To address this challenge, this study hypothesizes that unchanged regions between HRSI exhibit significantly more commonality features in the wavelet domain than changed regions. Based on this, this paper proposes an interactive wavelet commonality learning (IWCL) method for unsupervised end-to-end heterogeneous CD, which captures the wavelet commonality features between HRSI and identifies changes through their similarity measurement. The proposed IWCL consists of a Siamese wavelet encoder and two independent inverse wavelet decoders. Specifically, the wavelet encoder first decomposes the HRSI into multi-directional wavelet sub-bands, making the commonality features of the unchanged areas more prominent in multiple directions. Subsequently, we designed a cross-modal interactive learning (CMIL) module in the wavelet encoder to excavate deeper into these multi-directional wavelet commonality features. In addition, the wavelet commonality features extracted by the encoder are used to reconstruct the original image through two inverse wavelet decoders, which achieves unsupervised learning. Moreover, a multi-scale commonality loss function is introduced to effectively guide the network to further improve the representation ability of wavelet commonality features at different scales. Finally, the difference image is acquired by measuring the similarity of wavelet commonality features between HRSI, and the binary change image is obtained by Otsu. Experiments on four heterogeneous CD datasets verify the effectiveness and superiority of the proposed IWCL in resolving heterogeneous CD. The code will be available at https://github.com/TongfeiLiu/IWCL-for-MCD.

Remote sensing image change detection (RSICD) is a crucial technique for Earth observation. However, the mainstream RSICD methods still face two main challenges. First, the encoding stage often fails to capture fine-grained structural representations, particularly in scenarios involving cross-scale targets and complex boundaries. Second, the decoding stage lacks effective modeling of spectral heterogeneity and direction-sensitive channel interactions, which severely limits the ability to accurately recognize strip-shaped objects. To address these issues, this paper proposes a network for RSICD named FastSAM-CD, which extends FastSAM with two dedicated modules. First, we design a hyperfusion multi-view adapter (HFM Adapter) for the encoder of FastSAM-CD. It significantly enhances the model’s ability to capture fine-grained boundaries across scales by constructing multi-view paths in both spatial-channel and local-global dimensions. Second, we propose a spectral-axial dynamic modulation module (SDM Module) for the decoder. It enhances the decoder’s extraction of spatial-frequency domain information through axial sensing and frequency-domain analysis, significantly improving the model’s detection accuracy for strip-shaped objects. The experiments on three RSICD datasets demonstrate that the proposed method outperforms the existing mainstream RSICD methods in terms of accuracy and efficiency.

In recent years, deep learning algorithms have been widely regarded as the preferred method for remote sensing image analysis and have been successfully applied in the field of change detection. However, currently only a few comprehensive review papers on deep learning based remote sensing image change detection methods are available. To address the above issues, this study provides the current research status and development trends of remote sensing image change detection and analysis based on deep learning. Firstly, considering the differences in data volume and in data characteristics between different data sources, unlike previous reviews that only focus on the change detection problem of a certain type of remote sensing data, this review outlines the various data types involved in remote sensing image change detection, mainly including very high-resolution data, hyperspectral data, synthetic aperture radar data, and heterogeneous data. Secondly, unlike previous reviews that only introduce deep learning methods as a category of methods, this review comprehensively summarizes the research progress in remote sensing image change detection from three aspects: supervised deep learning, semi-supervised deep learning, and unsupervised deep learning, and explores the advantages and limitations of these methods. On this basis, we also propose five interesting research directions to promote further development in this field, including data privacy protection, multi-modality semantic-level change detection, lightweight models, change detection assisted by foundational models, as well as brain-inspired and vision-language models for change detection. This study will help deepen our understanding of deep learning in change detection in multiple ways and lay the foundation for future research.

Road change detection (RCD) is crucial for intelligent transportation, disaster assessment, and urban planning. However, current general change detection (CD) methods focus on various targets, such as buildings, while less attention is paid to the CD of narrow and elongated roads. Compared with general CD, RCD may still be limited by the following two aspects: On the one hand, the road usually occupies a small proportion of pixels in remote sensing images (RSIs) and is often easily blocked by buildings, trees, etc., making it difficult to ensure the integrity and connectivity of road structural features in RCD. On the other hand, RCD may easily be confused with the semantic information of similar material backgrounds (such as parking lots and building roofs) due to the lack of salient road features. To overcome the above limitations, we propose a skeleton-guided large-kernel attention Transformer (SLAFormer) for RCD, which can focus on salient road structural and semantic features to enhance its performance. In the proposed SLAFormer, we construct a novel skeleton-guided large-kernel attention module (SLKAM) and a frequency-guided cross spatial-channel difference module (FSCDM) to achieve the above goals. The SLKAM is used to make the model focus on road-specific skeleton features, which preserve the overall structure and morphology of roads to enhance the continuity and integrity of road features. In addition, the FSCDM is devised to better capture small-scale road changes and reduce semantic confusion with similar backgrounds, thereby enhancing change regions and extracting highly discriminative road difference information. Extensive experiments on two public RCD datasets show that ours achieves better RCD accuracy compared with several state-of-the-art (SOTA) approaches. The code will be available at https://github.com/TongfeiLiu/SLAFormer-for-RCD.

Multimodal change detection (MCD), as an emerging task, aims at recognizing change regions from bi-temporal remote sensing images (RSI) of different modalities. Inspired by the success of the self-attention mechanism in transformer, attempts have been made to solve MCD through the transformer variants. However, transformer-based network optimization requires high-quality training samples. In addition, due to the significant differences in the data distribution, semantic information, and feature representation of multimodal data, transformer-based methods have obvious deficiencies in local feature representation and spatial consistency, especially when dealing with heterogeneous images. To address the above challenges, we propose a collaborative frequency-aware transformer for MCD (CFAT-MCD). As an unsupervised framework, CFAT-MCD is capable of learning more fine-grained patterns of land cover change through a few pseudo-labels. The CFAT is designed to enhance spatial consistency and align the features on a multi-scale basis, which can effectively mitigate the effects of modal differences. In addition, we propose a window-based spatial-frequency collaborative representation (SFCR) module to introduce frequency information into the spatial domain and improve the discriminability of spatial features. Extensive experiments on public datasets and quantitative analyses have validated the superior detection performance of our approach and the effectiveness of each module.