Remote sensing (RS) change detection (CD) is a critical task in monitoring surface dynamics. Recently, Mamba-based methods have shown promising performance and are quickly adopted in change detection. However, When addressing the task of CD in complex scenarios, existing methods have limitations in capturing minor change features and texture change features due to the lack of frequency information. To address these challenges, we propose a frequency-enhanced Mamba for RSCD (FEMCD). First, we designed a difference-guided State Space Model (DGSSM) to extract change-related features. DGSSM takes the features from two images as input and uses absolute-difference features to guide the network to focus on change regions. Second, we developed a DCT-aided Mamba decoder (DCTMD) for feature decoding and refinement. DCTMD uses the Omnidirectional Selective Scan Module (OSSM) to refine the change-related features and DCT to capture minor change details. Finally, after undergoing the processing of a classifier, our network generates the ultimate change map. We conduct extensive experiments on five remote sensing change detection datasets, comparing FEMCD with eleven SOTA change detectors. The experimental results show that our proposed FEMCD method outperforms other compared methods. The code can be found at https://github.com/JYN712/FEMCD.

Most of the existing RGB-D saliency detectors have tried different strategies to fuse RGB and depth information to generate better saliency detection results. However, the ability of only using RGB image to detect the salient object is ignored in RGB-D saliency detectors. In this paper, we propose a triple-complementary network for fully exploring the RGB and depth information. The first two sub-networks are used for extracting saliency from RGB image and RGB-D image pair, respectively. The third sub-network refines the saliency map by comprehensively considering RGB image, depth map and the refined saliency map of the first two sub-networks. We also propose depth weighted refinement to suppress the high contrast objects with large depth values. The proposed method not only fully explores the information of RGB but also makes the RGB and depth tightly coupled. The experiments on five RGB-D saliency detection datasets show the superiority of the proposed method over the state-of-the-art RGB-D saliency detectors.

Change detection (CD) in remote sensing (RS) images serves as a vital method for identifying changes on the Earth's surface. Recent advancements in deep learning (DL)-based CD methods have shown considerable progress. However, there is still significant room for further improvement of CD performance, particularly in fine-grained detection, such as enhancing edge details and reducing pseudo-changes. To this end, a novel Multi-scale Edge Enhancement and Progressive Change-aware Network (MEPNet) is proposed to improve the ability of feature representation for changed objects. Specifically, we introduce an Edge Enhancement Module (EEM) to capture the long-range dependency, explicitly emphasizing high-frequency feature, and strengthening edge information to improve the accuracy of change regions. In addition, we propose a Progressive Change-aware Module (PCAM) that progressively applies depthwise separable convolutions with kernels of decreasing size to localize changes at different scales, enabling precise refinement of change objects and reducing pseudo-changes. These two components work together to advance the performance of MEPNet. Experimental results demonstrate that our method outperforms eleven SOTA methods on the LEVIR-CD, SYSU-CD, and CDD datasets, achieving superior accuracy and efficiency.

Change detection (CD) is a fundamental yet challenging problem, which aims at detecting changed object in two observations. Recent CD methods are designed based on the off‐the‐shelf semantic segmentation network architectures, which is not optimal for extracting and using change‐related features. In this paper, a novel CD network architecture is proposed, including change‐related feature extraction, cross feature enhancement, and multi‐level supervision. Absolute difference of the features of different convolutional layers is first computed from a Unet‐like network for two observations. The features are partitioned into high‐ and low‐level features according to their functionalities. Then the high‐ and low‐level features are recurrently refined by cross feature enhancement to increase the representational ability of the features. The network learns change‐related features with multi‐level supervisions. The final CD result can be obtained by fusing multiple predictions. Experimental results on three CD benchmark datasets indicate the superiority of the authors' method when compared with six state‐of‐the‐art deep learning‐based CD methods.