With the wide application of deep learning (DL), change detection (CD) for remote sensing images (RSIs) has realized the leap from the traditional to the intelligent methods. However, many existing methods still need further improvement in practical applications, especially in increasing the effectiveness of feature extraction and reducing model computational cost. In this paper, we propose a novel deep multiscale Siamese network with parallel convolutional structure (PCS) and self-attention (SA) (MSPSNet), which has excellent capabilities of feature extraction and feature integration under an acceptable consumption. It mainly contains three subnetworks: deep multiscale feature extraction, feature integration by the PCS, and feature refinement based on the SA. In the first subnetwork, a deep multiscale Siamese network based on convolutional block is designed to depict image features at different scales for different temporal images. In the subsequent subnetworks, a PCS model is proposed to integrate multiscale features of different temporal images, and then, a SA model is constructed to further enhance the representation of image information. Experiments are conducted on two public RSIs datasets, indicating that the proposed framework performs well in detecting changes.

The richness of spectral information in multitemporal hyperspectral images (MHSIs) offers the possibility to effectively detect subtle changes and properties of grounds. However, severe spectral variabilities and inadequate spatial co-exploitation capabilities limit the performance of existing methods due to differences in acquisition times and conditions. Therefore, this article proposes a strategy of joint unmixing and multitemporal spatial information coguidance (JUC) to fully exploit the spatio-temporal-spectral features. First, a multitemporal joint unmixing method is used to achieve endmembers’ extraction and abundance estimation. Wherein the method adds spectral perturbed regularization when compared to the traditional unmixing strategy, making it robust to spectral variability. Second, we propose a multitemporal coguidance method that highlights the contrast between changed and unchanged regions and removes the noise by transferring the common structure information between the multitemporal first principal component map and the abundance difference maps. It will obtain an enhanced abundance difference maps and achieve effective combination of multitemporal spatial information. The final change result can be obtained by combining and thresholding these enhanced abundance difference maps. Experiments on some data sets demonstrate that the proposed algorithm is sufficiently valid and robust for multitemporal images, especially for data containing spectral variabilities and obvious structures.

With the increase in the availability of multitemporal hyperspectral images (HSIs), HSIs change detection (CD) methods, including pixel-level and subpixel-level based methods, have attracted great attention in recent years. However, the widespread presence of mixed pixels in HSIs may make it difficult for pixel-level methods to detect subtle changes; meanwhile, the less utilization of spatial information may also lead to limitations in some subpixel-level methods. Therefore, a joint framework, which aims to combine the advantages of pixel-level in spatial utilization and subpixel-level in temporal and spectral exploration, is proposed to enhance the performance of HSIs CD. Two models, convolutional sparse analysis and temporal spectral unmixing, are introduced and presented to characterize different spatial structures and overcome the effects of spectral variability under this framework, respectively. In addition, a multiple CD-based on subpixel analysis is discussed as well. Experiments conducted on three bitemporal HSIs datasets indicate that the proposed framework is robust in capturing effective features and has achieved great detection accuracy.

Change detection (CD) for multitemporal remote sensing images can find change trends and reveal the development patterns. However, the typical methods based on algebraic operation, image transformation, or segmentation may not yield satisfactory results due to the spectral variability and noise complexity. In a sense, it can be considered that multitemporal images are composed of unchanged and changed regions, as well as noise. In the light of this, a stepwise subtraction method for CD is proposed, based on auto-updating multitemporal matrix factorization (AMMF) and spectral constraint, to remove the unchanged regions and noise from the original images step by step. The unchanged regions are first identified by AMMF, during which the distribution and subspace information of each temporal image are regularized to encode the spatio-temporal correlation. Then, mean shift smoothness is adopted as a spectral constraint so as to remove the noise. In this way, the changed regions have been highlighted so that the change map can be obtained by a postsegmentation method. Experiments have been conducted on three multitemporal data sets, including images from Quick Bird, aerial, and GF-1, indicating that the proposed method is effective and robust, which is superior to some state-of-the-art methods.

Deep learning-based methods have been widely applied to hyperspectral image change detection (HSI-CD), where most existing approaches adopt a patch-wise framework by taking paired patches from bi-temporal HSIs as input and determining whether the central position has changed. However, these methods suffer from several limitations. Being patch-wise, they fail to fully exploit the rich information contained in the entire bi-temporal HSIs and often overlook useful content outside the training samples. Although some methods incorporate advanced modeling techniques, their receptive fields are still constrained by the patch size, which limits their ability to capture global dependencies. Besides, many approaches underutilize the inherent advantage of HSI in capturing subtle spectral differences. To address these challenges, we propose a Mamba-based image-level global difference-aware network (GDAMamba) for HSI-CD, consisting of three key modules: the global contextual difference awareness module (GCDAM), the temporal spectral difference enhancement module (TSDEM), and the adaptive gated fusion module (AGFM). Specifically, GCDAM enables global spatial contextual awareness by establishing associations between each pixel and all others on the difference map. TSDEM processes the bi-temporal HSIs in a pixel-wise manner, progressively extracts and enhances spectral differences. These differences are then integrated into the global difference representations through AGFM, which are subsequently fed into a segmentation head to generate the final change detection map. Experimental results on three benchmark datasets demonstrate that GDAMamba outperforms several state-of-the-art methods. Our code is available at https://github.com/Preston-Dong/GDAMamba.

Semantic change detection (SCD) extends the binary change detection (BCD) task, as it not only locates change areas but also identifies change transition types. Recent research has verified that a multitask learning network performs well in tackling the SCD task, which jointly addresses the two subtasks of change localization and semantic identification. However, it remains a challenge to specifically optimize the distinct features of these two subtasks and establish their interaction to further enhance the overall performance of the multitask network efficiently. In this letter, we propose a novel SCD method that emphasizes the refinement and collaboration of difference and semantic features (RCDSFs). Specifically, a difference feature refinement branch (DFRB) is designed to integrate temporal information and highlight the difference features. Simultaneously, a semantic context refinement branch (SCRB) is developed to extract multiscale and cross-scale semantic details. Moreover, a simple yet effective feature interaction-fusion module (FIFM) is incorporated to coordinate the two subtasks, ensuring consistency while providing additional auxiliary information for each other. Comprehensive experiments on two public remote sensing image SCD datasets demonstrate that the proposed method outperforms the state-of-the-art algorithms. The code will be available at https://github.com/wanglinlin0219/RCDSF