SAM-Mamba: A Two-Stage Change Detection Network Combining the Adapting Segment Anything and Mamba Models

Remote sensing change detection (CD) involves identifying differences between two satellite images of the same geographic area taken at different times. It plays a critical role in applications such as urban planning and disaster management. Traditional CD methods rely on manually extracted features, which often lack robustness and accuracy in capturing the details of objects. Recently, deep learning-based methods have expanded the applications of CD in high-resolution remote sensing images, yet they struggle to fully utilize the multi-level features extracted by backbone networks, limiting their performance. To address this challenge, we propose a Semantic-Guided Cross-Attention Network (SCANet). It introduces a Hierarchical Semantic-Guided Fusion (HSF) module, which leverages high-level semantic information to guide low-level spatial details through an attention mechanism. Additionally, we design a Cross-Attention Feature Fusion (CAFF) module to establish global correlations between bitemporal images, thereby improving feature interaction. Extensive experiments on the IWHR-data and LEVIR-CD datasets demonstrate that SCANet significantly outperforms existing State-of-the-Art (SOTA) methods. Specifically, the F1-score and the Intersection over Union (IoU) score are improved by 2.002% and 3.297% on the IWHR-data dataset and by 0.761% and 1.276% on the LEVIR-CD dataset, respectively. These results validate the effectiveness of semantic-guided fusion and cross-attention feature interaction, providing new insights for advancing change detection research in high-resolution remote sensing imagery.

Change detection (CD) is an important remote sensing (RS) data analysis technology. Existing remote sensing change detection (RS-CD) technologies cannot fully consider situations where pixels between bitemporal images do not correspond well on a one-to-one basis due to factors such as seasonal changes and lighting conditions. Existing networks construct two identical feature extraction branches through convolution, which share weights. The two branches work independently and do not merge until the feature mapping is sent to the decoder head. This results in a lack of feature information interaction between the two images. So, directing attention to the change area is of research interest. In complex backgrounds, the loss of edge details is very important. Therefore, this paper proposes a new CD algorithm that extracts multi-scale feature information through the backbone network in the coding stage. According to the task characteristics of CD, two submodules (the Feature Interaction Module and Detail Feature Guidance Module) are designed to make the feature information between the bitemporal RS images fully interact. Thus, the edge details are restored to the greatest extent while fully paying attention to the change areas. Finally, in the decoding stage, the feature information of different levels is fully used for fusion and decoding operations. We build a new CD dataset to further verify and test the model’s performance. The generalization and robustness of the model are further verified by using two open datasets. However, due to the relatively simple construction of the model, it cannot handle the task of multi-classification CD well. Therefore, further research on multi-classification CD algorithms is recommended. Moreover, due to the high production cost of CD datasets and the difficulty in obtaining them in practical tasks, future research will look into semi-supervised or unsupervised related CD algorithms.

Remote sensing change detection (RSCD) is an important technique for extracting information about changes to the surface from bi-temporal remote sensing images. However, existing RSCD datasets are primarily based on satellite images, and those built from high-resolution UAV images are scarce. Therefore, this paper proposes a high-resolution, UAV-based change detection dataset (UAV-CD). In addition, UAV change detections differ significantly from satellite change detections. UAV images have higher spatial resolution and richer detailed information, but they are susceptible to interference from non-surface structural factors, such as shadows, illumination fluctuations, and vehicles. In contrast, satellite images provide wide coverage but lack sufficient detailed representation. Furthermore, most existing methods are specifically designed for satellite image characteristics, making them difficult to adapt to the high-detail and high-noise scenarios of UAV images. To overcome these challenges, we propose SFCF-Net. This introduces a Single-scale Global Feature Enhancement Module (SGFEM) to capture global contextual information from a single-scale perspective. The Multi-scale Dynamic Feature Integration Module (MDFIM) dynamically integrates semantic and structural information at different scales to effectively mitigate the problem of semantic inconsistency in single-scale features. The Multi-receptive Field Local Feature Fusion Module (MFLFM) aggregates local detail information through multi-scale convolutions to compensate for the lack of edge detail perception in global modelling. Experimental results on the UAV-CD, LEVIR-CD and CDD datasets demonstrate that SFCF-Net outperforms other baseline change detection networks, achieving detection accuracies of 92.73%, 92.95% and 90.84% respectively. The source code is available in:https://github.com/kwxt/SFCF-Net.git.

Natural reserves play a leading role in safeguarding national ecological security. Remote sensing change detection (CD) technology can identify the dynamic changes of land use and warn of ecological risks in natural reserves in a timely manner, which can provide technical support for the management of natural reserves. We propose a CD method (CA-BIT) based on the improved bitemporal image transformer (BIT) model to realize the change detection of remote sensing data of Anhui Natural Reserves in 2018 and 2021. Resnet34-CA is constructed through the combination of Resnet34 and a coordinate attention mechanism to effectively extract high-level semantic features. The BIT module is also used to efficiently enhance the original semantic features. Compared with the overall accuracy of the existing deep learning-based CD methods, that of CA-BIT is 98.34% on the natural protected area CD datasets and 99.05% on LEVIR_CD. Our method can effectively satisfy the need of CD of different land categories such as construction land, farmland, and forest land.

Change detection (CD) in remote sensing imagery is crucial for monitoring environmental variations over time. Recent advancements have introduced numerous effective algorithms, significantly enhancing the performance of CD tasks. However, various challenges still persist, such as complex backgrounds, small targets, and diverse change types. To address these challenges, we propose a novel Mamba-based multi-scale feature enhancement method. Our method leverages the strengths of state space models in handling long sequences by reformulating the pyramid (multi-scale) feature fusion as a sequence processing task. This approach enables our model to efficiently capture both long-range context and fine-grained local details, which are critical for effective feature learning and mitigating the aforementioned challenges. Moreover, we introduce scale-level embedding and auxiliary loss to optimize and regulate the model's training process. These enhancements not only accelerate training but also improve the model's generalization capability. Extensive experiments are conducted on four leading benchmarks: WHUCD, SYSU-CD, LEVIR-CD, and DSIFN-CD datasets. The results demonstrate that our approach consistently delivers state-of-theart performance, surpassing other cutting-edge methods by a significant margin. The source code will be made available at https://github.com/Band-127/Pyramid-Mamba-CD upon formal acceptance of the paper.

This paper investigates the influence of depth of field on change detection in both pictorial and solid scenes. In this work, a within-subjects experiment is conducted using a flicker paradigm, with which the hit rate and response time for change detection are obtained. The results show that depth of field has effects on change detection: the hit rate is smaller and response time is longer in the scene with small depth of field than in the scene with large depth of field or uniform blur. It is concluded that when depth of field is small and binocular disparity is not zero in a picture, the influence of depth of field on change detection is more significant than binocular disparity. This conclusion leads to the result that the change in the sharp area is detected easier and faster than in the area that is closer to the observer.

Timely and accurate change information in the urban environment is essential for successful planning and management. The change detection may range from 1) monitoring general land cover/land use found in multiple dates of imagery, to 2) anomaly (e.g., subsidence) detection on hazardous waste sites. Remote sensing approaches to change detection have been widely used due to its cost-effectiveness, extensibility, and temporal frequency. Since the advent of high-spatial resolution satellite imagery, it has become increasing popular to detect, analyze, and monitor detailed changes such as new buildings, roads, and even patios in the urban environment. Basically, there are two types of change detection methods: 1) detection of the change using various image enhancement methods, and 2) extraction of detailed types of land-cover change based on the use of classification techniques (Chan et al. 2001; Jensen 2005)

Object-oriented change detection (OOCD) plays an important role in remote sensing change detection. Generally, most of current OOCD methods adopt the highest predicted probability to determine whether objects have changes. However, it ignores the fact that only parts of an object have changes, which will generate the uncertain classification information. To reduce the classification uncertainty, an improved rough-fuzzy possibilistic <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$c$</tex-math></inline-formula> -means clustering algorithm combined with multiresolution scales information (MRFPCM) is proposed. First, stacked bitemporal images are segmented using the multiresolution segmentation approach from coarse to fine scale. Second, objects at the coarsest scale are classified into changed, unchanged, and uncertain categories by the proposed MRFPCM. Third, all the changed and unchanged objects in previous scales are combined as training samples to classify the uncertain objects into new changed, unchanged, and uncertain objects. Finally, segmented objects are classified layer by layer based on the MRFPCM until there are no uncertain objects. The MRFPCM method is validated on three datasets with different land change complexity and compared with five widely used change detection methods. The experimental results demonstrate the effectiveness and stability of the proposed approach.

Remote sensing change detection (CD) is a highly significant subtask within the field of Earth observation. Recently, weakly supervised CD (WSCD) methods based on image-level annotations have attracted interest, it is challenging to generate a clear margin between changed and unchanged regions with a lack of detailed annotation. In this article, based on class activation maps (CAMs), we propose a novel WSCD network based on self-supervised learning and change mask guidance (SSCMNet). First, we design a multiscale self-supervised constraint (MSC) module to narrow the gap between weak supervision and full supervision and compensate for the inherent shortcomings of CAMs. Second, a change mask guidance (CMG) module is proposed to further guide the network to keep the integrity of changed objects according to the consistency within unchanged regions and inconsistency within changed regions. Finally, to address the challenge of transferring commonly used post-processing methods in semantic segmentation to CD, an adaptive post-processing (APP) module is designed to adaptively select one of the input images for post-processing. We conduct experiments on three publicly available remote sensing CD datasets. Quantitative metrics and visualized results demonstrate the outstanding performance of the proposed method.

Remote Sensing (RS) images capture spatial-temporal data on the Earth's surface that is valuable for understanding geographical changes over time. Change detection (CD) is applied in monitoring land use patterns, urban development, evaluating disaster impacts among other applications. Traditional CD methods often face challenges in distinguishing between changes and irrelevant variations in data, arising from comparison of pixel values, without considering their context. Deep feature based methods have shown promise due to their content extraction capabilities. However, feature extraction alone might not be enough for accurate CD. This study proposes incorporating spatial-temporal dependencies to create contextual understanding by modelling relationships between images in space and time dimensions. The proposed model processes dual time points using parallel encoders, extracting highly representative deep features independently. The encodings from the dual time points are then concatenated and passed through Long Short Term Memory (LTSM) layers and a decoder. The output from the LSTM is then concatenated with that from the decoder in a space-time feature fusion. This optimizes information representation of spectral, spatial and temporal details, in RS images before further analysis of change. This approach aims to maximize authentic information while reducing noise interference by introducing the context of change. Compared to conventional CD methods, the proposed technique achieves higher overall accuracy of 97.4%, an F1 Score of 89% and an intersection over union (IoU) of 86.7 when evaluated on the EGY-BCD dataset. The results demonstrate the potential of incorporating spatial-temporal dependencies in CD tasks for RS images.

Facing the challenge of small open labeled data sets in remote sensing change detection, this letter proposes a novel supervised change detection method by taking advantages of capsule network which can reach the same performance as traditional convolutional neural networks (CNNs) but with less training data. To achieve this aim, we propose a pseudo-Siamese capsule network which takes both rotational invariance and spatial hierarchies between features into account for aerial images change detection. First, the features of image pairs are extracted by two identical nonshared weights convolutional capsule networks. Second, the extracted features are directly concatenated and sent to another convolutional capsule layer. The change probability map is obtained by calculating the length of the capsule vectors in the final layer. Additionally, to reduce the influence of imbalance samples when we optimize our network, we design a margin-focal loss function to pay more attention to the misclassified samples. Finally, binary change map can be produced by a simple threshold. Experimental results carried out on the SZTAKI AirChange Benchmark Set show that the proposed method achieves comparable and even better results with existing state-of-the-art methods in terms of F-measure.

Currently, deep learning networks based on architectures such as CNN and Transformer have achieved significant advances in remote sensing image change detection, effectively addressing the issue of false changes due to spectral and radiometric discrepancies. However, when handling remote sensing image data from multiple sensors, different viewing angles, and extended periods, these models show limitations in modelling dynamic interactions and feature representations in change regions, restricting their ability to model the integrity and precision of irregular change areas. We propose the Context-Aware Global-Local Subspace Attention Change Detection Network (CGLCS-Net) to resolve these issues and introduce the Global-Local Context-Aware Selector (GLCAS) and the Subspace-based Self-Attention Fusion (SSAF) module. GLCAS dynamically selects receptive fields at different feature extraction stages through a joint pooling attention mechanism and depthwise separable convolution, enhancing global context and local feature extraction capabilities and improving feature representation for multi-scale and irregular change regions. The SSAF module establishes dynamic interactions between dual-temporal features via feature decomposition and self-attention mechanisms, focusing on semantic change areas to address challenges such as sensor viewpoint variations and the texture and spectral inconsistencies caused by long periods. Compared to ChangeFormer, CGLCS-Net achieved improvements in the IoU metric of 0.95%, 9.23%, and 13.16% on the three public datasets, i.e., LEVIR-CD, SYSU-CD, and S2Looking, respectively. Additionally, it reduced model parameters by 70.05%, floating-point operations by 7.5%, and inference time by 11.5%. These improvements enhance its applicability for continuous land use and land cover change monitoring.

Generative adversarial networks (GANs) possess powerful image translation capabilities. They can transform images acquired from different sensors into a unified domain, effectively mitigating the incomparability problem caused by imaging discrepancies in multimodal remote sensing change detection (CD). However, existing approaches predominantly emphasize domain unification while neglecting the loss of fine-grained features inherent in the translation process, consequently compromising both image translation quality and CD accuracy. To overcome these limitations, we propose a novel texture and structure interaction guided GAN (TSIG-GAN). This network establishes interactive guidance between image texture and structural features through a carefully designed dual-stream cross encoder-decoder architecture, enabling in-depth mining of fine-grained features and significantly improving the fidelity of translated images. Furthermore, to address the spatial scale diversity and complexity of remote sensing images, we develop a multi-scale adaptive feature pyramid (MAFP) module and a contextual semantic interaction guidance (CSIG) mechanism, aiming to further strengthen the model's robust representation of fine-grained features across multiple scales and complex scenes. Specifically, the MAFP module effectively captures spatial details of targets at different resolutions by dynamically integrating multi-scale features, thereby preventing detail loss in small objects due to scale discrepancies. The CSIG mechanism achieves deep interaction between texture and structural features at the contextual semantic level, further promoting their mutual cooperation, thereby enhancing the consistency of fine-grained features representation and semantic integrity in complex scenes. Finally, the translated fine-grained images are fed into a custom CD network to extract changes. To evaluate the effectiveness of the proposed method, we conducted systematic experiments on five representative real-world datasets and performed comparative analysis with sixteen state-of-the-art multimodal CD methods. The experimental results demonstrate that TSIG-GAN achieves significant improvements in both image translation and CD performance, exhibiting superior fine-grained restoration capability and change identification capability.

Achieving accurate bladder wall and tumor segmentation from MRI is critical for diagnosing and treating bladder cancer. However, automated segmentation remains challenging due to factors such as comparable density distributions, intricate tumor morphologies, and unclear boundaries. Considering the attributes of bladder MRI images, we propose an efficient multiscale hierarchical hybrid attention with a transformer (MH2AFormer) for bladder cancer and wall segmentation. Specifically, a multiscale hybrid attention and transformer (MHAT) module in the encoder is designed to adaptively extract and aggregate multiscale hybrid feature representations from the input image. In the decoder stage, we devise a multiscale hybrid attention (MHA) module to generate high-quality segmentation results from multiscale hybrid features. Combining these modules enhances the feature representation and guides the model to focus on tumor and wall regions, which helps to solve bladder image segmentation challenges. Moreover, MHAT utilizes the Fast Fourier Transformer with a large kernel (e.g., 224 × 224) to model global feature relationships while reducing computational complexity in the encoding stage. The model performance was evaluated on two datasets. As a result, the model achieves relatively best results regarding the intersection over union (IoU) and dice similarity coefficient (DSC) on both datasets (Dataset A: IoU = 80.26%, DSC = 88.20%; Dataset B: IoU = 89.74%, DSC = 94.48%). These advantageous outcomes substantiate the practical utility of our approach, highlighting its potential to alleviate the workload of radiologists when applied in clinical settings.

Remote sensing change detection (RSCD) is a key tool for environmental monitoring and resource management, playing a significant role in monitoring dynamic surface changes. In practical applications, RSCD often requires high precision and efficient detection methods. However, traditional methods tend to involve high technical complexity and a large number of parameters, and are susceptible to interference from complex background noise, leading to poor performance in detecting change areas. To address these issues, this paper proposes a lightweight remote sensing change detection network, LMG-Net. The model uses a lightweight encoder and incorporates a Hierarchical Transformer Module (HTF) to suppress background noise and minimize parameter increase, effectively extracting multi-level global features. Additionally, the paper introduces a Multi-dimensional Cooperative Attention Guidance (MAG) mechanism, further enhancing the ability to detect boundary changes. The model has only 3.29M parameters and a computational load of 3.89G, demonstrating its high applicability, particularly for real-time applications in resource-constrained environments. Experimental results show that LMG-Net achieves state-of-the-art F1 scores and IoU values on the WHU-CD, SYSU-CD, and LEVIR-CD+ datasets: (94.79%, 90.09%), (82.29%, 69.90%), and (84.30%, 71.14%).