MAEMACD: a MAE-enhanced multiresolution attention network for remote sensing image change detection

Change detection is a crucial technique for identifying change information between image pairs of the same geographical area. Existing deep learning-based change detection methods achieve high performance by utilizing annotated and registered bi-temporal images. However, obtaining accurate annotations and registered bi-temporal images requires expert knowledge and substantial financial resources. The effectiveness of models may be constrained by the limited variation scenarios within available change detection datasets. To address these issues, we develop a novel single-temporal self-supervised learning framework for change detection, namely S3FCD, which facilitates training a bi-temporal remote sensing image change detection model without labeled and registered paired data. Specifically, a coarse change generator is first employed to generate pairs of training data for training a preliminary change detection model. To improve the quality of the generated image pairs, a deep feature-based generator (DFG) module is designed based on the pre-trained model. To enhance the diversity of the generated image pairs, a patch memory bank (PMB) module is integrated into DFG for storing and managing patches. S3FCD has demonstrated state-of-the-art performance across multiple datasets.

Change detection (CD) is essential in remote sensing (RS) for natural resource monitoring, territorial planning, and disaster assessment. With the abundance of data collected by satellite, aircraft, and unmanned aerial vehicles, the utilization of multisource RS image CD (RSICD) enables the efficient acquisition of ground object change information and timely updates to existing databases. Although CD techniques have been developed and successfully applied for approximately six decades, a systematic and comprehensive review that addresses emerging trends, including multisource, data-driven, and large-scale artificial intelligence (AI) models, is lacking. Therefore, first, the development process of RSICD was reviewed. Second, the characteristics of multisource RS images were analyzed, and all publicly available RSICD data that we could gather were collected and organized. Third, RSICD methods were systematically classified and summarized on the basis of the detection framework, detection granularity, and data sources. Fourth, the suitability of specific data and CD methods for diverse applications and tasks was assessed. Finally, challenges, opportunities, and future directions for RSICD were discussed within the context of high-resolution imagery, multisource data, and large-scale AI models. This review can help researchers better understand this field, shed light on this topic, and inspire further RSICD research efforts.

Remote sensing image change detection (RSICD) is an essential measure for monitoring the earth’s surface changes. In recent years, the explosive growth of very high-resolution (VHR) satellite sensors and the booming innovations in deep learning technology have significantly boosted RSICD development. However, most of the current RSICD models focus on locating accurate change areas while ignoring the efficiency of their method, which limits the practical application of RSICD models, especially for large-scale and emergency RSICD tasks. In this paper, we propose an Efficient Multi-scale-fusion Change Detection Network (EMS-CDNet) for bi-temporal RSICD tasks. Our EMS-CDNet pays more attention to the model’s inference speed and the accuracy-efficiency trade-off rather than only pursuing detection accuracy. We designed a multi-scale fusion module for EMS-CDNet, which adopts multi-scale and multi-branch operations to extract multi-scale features simultaneously and aggregate features at different feature levels. In addition to EMS-CDNet’s ability to achieve sufficient feature extraction, the multi-scale image input within the designed module alleviates the influence of image registration errors in practical applications, thereby strengthening EMS-CDNet’s value for practical RSICD tasks. We also integrated a novel partition unit in EMS-CDNet to lighten the model while maintaining the detection ability of small targets, thus shortening its processing time without a severe accuracy decrease. We conducted experiments on two state-of-the-art (SOTA) public RSICD datasets and our own collected dataset. The public datasets were utilized to comparatively measure the overall accuracy and efficiency measurement of EMS-CDNet, and the dataset of images we collected was used to observe EMS-CDNet’s performance under the influence of image registration errors. Our experimental results show that EMS-CDNet achieved a better accuracy-efficiency trade-off than the SOTA public datasets methods. For example, EMS-CDNet reduced the inference time by about 33% while maintaining identical detection accuracy to CLNet (the optimal method among the comparison methods). Furthermore, EMS-CDNet achieved higher accuracy on our collected dataset, with an F1 of 74% and mIoU of 0.806, demonstrating its robustness to image registration errors and showing its value for practical RSICD applications.

ABSTRACTIn this article, we propose a novel difference image (DI) creation method for unsupervised change detection in multi-temporal multi-spectral remote-sensing images based on deep learning theory. First, we apply deep belief network to learn local and high-level features from the local neighbour of a given pixel in an unsupervised manner. Second, a back propagation algorithm is improved to build a DI based on selected training samples, which can highlight the difference on changed regions and suppress the false changes on unchanged regions. Finally, we get the change trajectory map using simple clustering analysis. The proposed scheme is tested on three remote-sensing data sets. Qualitative and quantitative evaluations show its superior performance compared to the traditional pixel-level and texture-level-based approaches.

With the rapid development of remote sensing technologies, the frequency of observations of the same location is increasing, and many satellites and sensors produced a large amount of time series images. These images make long-term change detection and dynamic characteristic estimation of ground features possible. However, conventional remote sensing image change detection methods mostly rely on manual visual interpretation and supervised or unsupervised computer-aided classification. Traditional methods always face many bottlenecks when processing big and fast-growing datasets, such as low computational efficiency, low level of automation, and different identification standards and accuracies caused by different operators. With the rapid accumulation of remote sensing data, it has become an important but more challenging task to conduct change detection in a more precise, automated and standardized way. The development of geointelligent computing technologies provides a means of solving these problems and improve the accuracy and efficiency of remote sensing image change detection. In this paper, we presented a novel deep learning model called nest network(NestNet) based on a convolutional neural network to improve the accuracy of the automatic change detection task by using remotely sensed time series images. NestNet extracts the respective features of bi-temporal images using an encoding parallel module and subsequently employs absolute different operations to process the features of two images. Compared with change detection method based on U-Shaped network plus plus (UNet++), the parallel module improves the efficiency of NestNet. Finally, a decoding module is used to generate a predicted change image. This paper compares NestNet to traditional methods and state-of-the-art deep learning models on two datasets. The experimental results demonstrate that the accuracy of NestNet is better than that of state-of-the-art methods. It can be concluded that the NestNet model is a potential approach for change detection using high resolution remote sensing images.

ADS-Net:An Attention-Based deeply supervised network for remote sensing image change detection

With the increase of spatial resolution of remote sensing images, features of feature imaging become more and more complex, and the change detection methods based on techniques such as texture representation and local semantics are difficult to meet the demand. Most change detection methods usually focus on extracting semantic features and ignore the importance of high-resolution shallow information and fine-grained features, which often lead to uncertainty in edge detection and small target detection. For single-input networks when two temporal images are connected, the shallow layer of the network cannot provide the information of the individual original image to the deep layer features to help reconstruct the image, and therefore, the change detection results may be missing in detail and feature compactness. For this purpose, a twins context aggregation network (TCANet) is proposed to perform change detection on remote sensing images. In order to reduce the loss of spatial accuracy of remote sensing images and maintain high-resolution representation, we introduce HRNet as our backbone network to initially extract the features of interest. Our proposed context aggregation module (CAM) can amplify the convolutional neural network receptive field to obtain more detailed contextual information without significantly increasing the computational effort. The side output embedding module (SOEM) is proposed to improve the accuracy of small volume target change detection as well as to shorten the training process and speed up the detection while ensuring the performance. The method has experimented on the publicly available CDD dataset, the SYSU-CD dataset, and a challenging DSIFN dataset. With significant improvements in precision, recall, F1 score, and overall accuracy, the method outperforms the five methods mentioned in the literature.

Remote sensing image change detection is the detection process of determining the surface change area and change feature type for the same image area from multiple time series remote sensing data. It is the core technical means of land use change detection and land cover change detection, and it is also the key research field of remote sensing applied science. In view of the problems of time-consuming, labor-intensive, and low detection accuracy in the past image change detection methods, researchers in related fields have proposed more and more cutting-edge remote sensing image change detection methods. This article first describes the development of remote sensing image change detection methods, then explains the conventional processing flow of change detection and the conventional methods of change monitoring, and finally discusses the research progress of more mainstream change detection methods.

VHR imagery change detection is one of research hotspots and difficulties in the field of remote sensing. However, the traditional remote sensing image change detection method is a waste of time and energy and low efficiency. In recent years, deep learning approaches in remote sensing image change detection verified feasible and save time to improve efficiency. A UNet change detection method based on aggregation residuals and attention mechanism is proposed, using prior knowledge of deep learning. The UNet model is used as the basic model, and the aggregation residual module is introduced in the up-down sampling stage, which can fully extract the feature information of the image. The weight of each component in the feature graph can be adjusted by adding attention module in the jump connection layer. In the process of experiment based on the model parameters are reasonable and effective set of data sets to Longnan remote sensing image change detection, and the experimental results showing that compared with the traditional deep learning semantic segmentation method, this article methods F1 value of 0.873, the generated change detection figure closer to label figure, higher accuracy, shorter.

With the remarkable success of change detection (CD) in remote sensing images in the context of deep learning, many convolutional neural network (CNN) based methods have been proposed. In the current research, to obtain a better context modeling method for remote sensing images and to capture more spatiotemporal characteristics, several attention-based methods and transformer (TR)-based methods have been proposed. Recent research has also continued to innovate on TR-based methods, and many new methods have been proposed. Most of them require a huge number of calculation to achieve good results. Therefore, using the TR-based mehtod while maintaining the overhead low is a problem to be solved. Here, we propose a GNN-based multi-scale transformer siamese network for remote sensing image change detection (GMTS) that maintains a low network overhead while effectively modeling context in the spatiotemporal domain. We also design a novel hybrid backbone to extract features. Compared with the current CNN backbone, our backbone network has a lower overhead and achieves better results. Further, we use high/low frequency (HiLo) attention to extract more detailed local features and the multi-scale pooling pyramid transformer (MPPT) module to focus on more global features respectively. Finally, we leverage the context modeling capabilities of TR in the spatiotemporal domain to optimize the extracted features. We have a relatively low number of parameters compared to that required by current TR-based methods and achieve a good effect improvement, which provides a good balance between efficiency and performance.

Although some progress has been made in deep learning-based remote sensing image change detection, the complexity of scenes and the diversity of changes in remote sensing images lead to challenges related to background interference. For instance, remote sensing images typically contain numerous background regions, while the actual change regions constitute only a small proportion of the overall image. To address these challenges in remote sensing image change detection, this paper proposes a Dynamic Adaptive Context Attention Network (DACA-Net) based on an exchanging dual encoder–decoder (EDED) architecture. The core innovation of DACA-Net is the development of a novel Dynamic Adaptive Context Attention Module (DACAM), which learns attention weights and automatically adjusts the appropriate scale according to the features present in remote sensing images. By fusing multi-scale contextual features, DACAM effectively captures information regarding changes within these images. In addition, DACA-Net adopts an EDED architectural design, where the conventional convolutional modules in the EDED framework are replaced by DACAM modules. Unlike the original EDED architecture, DACAM modules are embedded after each encoder unit, enabling dynamic recalibration of T1/T2 features and cross-temporal information interaction. This design facilitates the capture of fine-grained change features at multiple scales. This architecture not only facilitates the extraction of discriminative features but also promotes a form of structural symmetry in the processing pipeline, contributing to more balanced and consistent feature representations. To validate the applicability of our proposed method in real-world scenarios, we constructed an Unmanned Aerial Vehicle (UAV) remote sensing dataset named the Guangxi Beihai Coast Nature Reserves (GBCNR). Extensive experiments conducted on three public datasets and our GBCNR dataset demonstrate that the proposed DACA-Net achieves strong performance across various evaluation metrics. For example, it attains an F1 score (F1) of 72.04% and a precision(P) of 66.59% on the GBCNR dataset, representing improvements of 3.94% and 4.72% over state-of-the-art methods such as semantic guidance and spatial localization network (SGSLN) and bi-temporal image Transformer (BIT), respectively. These results verify that the proposed network significantly enhances the ability to detect critical change regions and improves generalization performance.

Image change detection is one of the important application branches of remote sensing technology in many fields. However, in complex environments, remote sensing image change detection is often subject to various interferences, resulting in low accuracy and poor real-time performance of detection results. The research focuses on the advantages and problems of residual networks and depth-wise separable convolution modules, designs a new remote sensing image change detection model, and finally sets up experiments for verification. The average accuracy of the proposed detection model before and after training convergence was 0.54 and 0.97. The accuracy of repeated detection ranged from 95.82% to 99.68%, and the area under curve of the model was 0.90. However, after removing the integrated residual attention unit and depth-wise separable convolution, the accuracy decreased by 1.91% and the latency increased by 117ms. In addition, the detection efficiency of the model for different elements ranged from 0.91 to 0.94, with high accuracy in detecting changes in spatial and temporal scales and small offsets. The actual accuracy and mean latency time of the model were 92.43% and 260ms, respectively. In summary, the proposed change detection model significantly improves the accuracy and real-time performance of remote sensing image processing, contributing to the expanded application of remote sensing dynamic detection technology in fields such as ocean monitoring and ecological research.

Remote sensing image change detection techniques are widely used in environmental change detection such as landuse change monitor, flood monitor. Many change detection techniques are used in practice today. This paper reports the development of techniques based on artificial neural networks and presents a new method of integrating artificial neural networks (ANN) with gray system theory for remote sensing image change detection. Gray system theory, founded by Professor Deng Julong, can handle undetermined problem .It is effective when the sample datum can not satisfy some distribution. The accuracy of image change detection based on traditional ANN is influenced by some factors such as network architecture, training set. The number of hidden layers and the number of nodes in a hidden layer are not easy to deduce. The traditional neural network architecture which gives the best results for image change detection can only be determined experimentally, and this can be a lengthy process especially for large image. This paper presents a new method that the number of nodes in hidden layers is deduced by using gray correlation analysis in gray system theory. A neural network based change detection system using the backpropagation training is developed. The trained three-layered neural network was able to provide information of changes and detect land-cover change with an overall accuracy of 91.3 percent. Using the same training data, a maximum-likelihood supervised classification produced an accuracy of 85.1 percent. The experimental results by using multitemporal TM imagery and SPOT imagery. Findings of this study demonstrated the potential and advantages of using neural network and gray system theory in multitemporal change analysis.

Remote sensing image change detection is a core task of remote sensing image analysis; its purpose is to identify and quantify land cover changes in different periods. However, when the existing methods deal with complex features and subtle changes in buildings, vegetation, water bodies, roads, and other ground objects, there are often problems of false detection and missing detection, which affect the detection accuracy. To improve the accuracy of change detection, a multi-scale feature fusion network based on difference enhancement (FEDNet) is proposed. The FEDNet consists of a difference enhancement module (DEM) and a multi-scale feature fusion module (MFM). By summing the variation features of two-phase remote sensing images, the DEM enhances pixel-level differences, captures subtle changes, and aggregates features. The MFM fully integrates the multi-stage deep semantic information, which enables better extraction of changing features in complex scenes. Experiments on the LEVIR-CD, CLCD, WHU, NJDS, and GBCNR datasets show that the FEDNet significantly improves the detection efficiency of changes in buildings, cities, and vegetation. In terms of F1 value, IoU (Intersection over Union), precision, and recall rate, the FEDNet is superior to existing methods, which verifies its excellent performance.

Remote sensing image change detection method plays a great role in land cover research, disaster assessment, medical diagnosis, video surveillance, and other fields, so it has attracted wide attention. Based on a small sample dataset from SZTAKI AirChange Benchmark Set, in order to solve the problem that the deep learning network needs a large number of samples, this work first uses nongenerative sample augmentation method and generative sample augmentation method based on deep convolutional generative adversarial networks, and then, constructs a remote sensing image change detection model based on an improved DeepLabv3+ network. This model can realize end-to-end training and prediction of remote sensing image change detection with subpixel convolution. Finally, Landsat 8, Google Earth, and Onera satellite change detection datasets are used to verify the generalization performance of this network. The experimental results show that the improved network accuracy is 95.1% and the generalization performance is acceptable.