Image Change Detection Research Articles

Hypergraph Self-Supervised Learning based Joint Spectral-Spatial-Temporal Feature Representation for Hyperspectral Image Change Detection

Hypergraph Self-Supervised Learning based Joint Spectral-Spatial-Temporal Feature Representation for Hyperspectral Image Change Detection

End-to-End Hyperspectral Image Change Detection Based on Band Selection

End-to-End Hyperspectral Image Change Detection Based on Band Selection

BiFA: Remote Sensing Image Change Detection With Bitemporal Feature Alignment

BiFA: Remote Sensing Image Change Detection With Bitemporal Feature Alignment

Wavelet Siamese Network With Semi-Supervised Domain Adaptation for Remote Sensing Image Change Detection

Wavelet Siamese Network With Semi-Supervised Domain Adaptation for Remote Sensing Image Change Detection

Uncertainty-Aware Graph Self-Supervised Learning for Hyperspectral Image Change Detection

Uncertainty-Aware Graph Self-Supervised Learning for Hyperspectral Image Change Detection

SAR Image Change Detection via Heterogeneous Graph With Multiorder and Multilevel Connections

SAR Image Change Detection via Heterogeneous Graph With Multiorder and Multilevel Connections

Assessing large-scale roadside tree removal using aerial imagery and crash analysis: A difference-in-differences approach

Transport agencies worldwide remove roadside trees in the name of traffic safety. In 2017, the Georgia Department of Transportation (GDOT), the agency managing state and federal highways in the US state of Georgia, began clear-cutting trees on all transport agency property statewide, amounting to thousands of hectares of cleared land. As justification, GDOT cited fatal traffic crashes with trees.In this paper, we quantified the extent of tree removal within a 60 m buffer along the five major interstates in Georgia, using aerial imagery from the United States Department of Agriculture (USDA) Farm Service Agency’s National Agricultural Imagery Project (NAIP). We used supervised image classification and change detection on aerial imagery between 2015 and 2021. We also collected pre- and post-treatment crash data and conducted a difference-in-differences analysis on sampled road segments to isolate the effect of tree removal on crash rates.Results showed that 28.4 sq km of previously forested land in the interstate corridors were cleared of trees, 25% of the total forested land in the corridors. Tree crash fatalities did not decrease.Findings provide insight into approaches to roadside safety like tree removal while helping assess impacts and plan for next steps. The lack of evidence for reduced tree crash fatalities indicate that transport agencies should consider measures other than large-scale tree removal to prevent fatal crashes with trees.

EGMT-CD: Edge-Guided Multimodal Transformers Change Detection from Satellite and Aerial Images

Change detection from heterogeneous satellite and aerial images plays a progressively important role in many fields, including disaster assessment, urban construction, and land use monitoring. Currently, researchers have mainly devoted their attention to change detection using homologous image pairs and achieved many remarkable results. It is sometimes necessary to use heterogeneous images for change detection in practical scenarios due to missing images, emergency situations, and cloud and fog occlusion. However, heterogeneous change detection still faces great challenges, especially using satellite and aerial images. The main challenges in satellite and aerial image change detection are related to the resolution gap and blurred edge. Previous studies used interpolation or shallow feature alignment before traditional homologous change detection methods, which ignored the high-level feature interaction and edge information. Therefore, we propose a new heterogeneous change detection model based on multimodal transformers combined with edge guidance. In order to alleviate the resolution gap between satellite and aerial images, we design an improved spatially aligned transformer (SP-T) with a sub-pixel module to align the satellite features to the same size of the aerial ones supervised by a token loss. Moreover, we introduce an edge detection branch to guide change features using the object edge with an auxiliary edge-change loss. Finally, we conduct considerable experiments to verify the effectiveness and superiority of our proposed model (EGMT-CD) on a new satellite–aerial heterogeneous change dataset, named SACD. The experiments show that our method (EGMT-CD) outperforms many previously superior change detection methods and fully demonstrates its potential in heterogeneous change detection from satellite–aerial images.

Transformer-based network with difference information aware attention module

Existing great progress has been made through remote sensing image change detection (CD) methods, which use convolutions and transformations appropriately. However, due to the complexity of the objects photographed on satellites, the out-of-phase of the same object may exhibit a fine feature diversity and a wide range of variations. Meanwhile, in the previous work, pure convolutional networks and Transformer networks that focus on local information perform better on non-local information. Therefore, current methods still have the problem of insufficient discrimination between irregular changes and diverse boundaries. Here, to solve this biphasic dilemma, we put forward a network that uses the difference information between bitemporal images as a guide to refine features. We use one feature map to subtract another and use multi-scale convolution to refine the difference map. We then designed an attention module similar to self-attention using differential information. In addition, considering the need for both local and global information in change detection, we use the full Swin Transformer framework with shift window self-attention to pass information between local areas. In addition, in order to enhance the network’s ability to recognize, members of our team reformulate supervised, traditional saliency map prediction through saliency and edge prediction. Notably, our proposed model significantly outperforms existing single-branch methods and state-of-the-art performance was achieved on four famous public datasets in terms of the F1-score index, interest in joint indices, and population indices.

Optical and Thermal Image Processing for Monitoring Rainfall Triggered Shallow Landslides: Insights from Analogue Laboratory Experiments

This study explores the innovative use of digital image processing (DIP) techniques, also named PhotoMonitoring, for analysing the triggering conditions of shallow landslides. The approach, based on the combination of optical and infrared thermographic imaging (IRT), was applied to a laboratory-scale slope, reproduced in a flume test apparatus. Three experiments were conducted to replicate rainfall-induced shallow landslides, applying change detection and digital image correlation analysis to both optical and thermal images. The method combines IRT’s ability to measure ground surface temperature changes with DIP’s capacity to track movement and displacement. Results showed the high reliability of the displacement time-series obtained through IRT-DIP with respect to the reference optical-DIP. The IRT-DIP technique also detects anomaly signals two minutes before landslide occurrence that can be regarded as a possible failure precursor. This study testifies to the potential of image analysis as a remote sensing technique, demonstrating the ability of DIP to capture the dynamics of shallow landslides, as well as the advantages of optical–IRT combinations to follow slope deformation processes during night-time. This approach, if properly adapted to real-scale scenarios, may contribute to a better understanding of landslide behaviour, improve landslide monitoring strategies, and promote more effective early warning systems (EWS).

A single-building damage detection model based on multi-feature fusion: A case study in Yangbi

Accurate and effective identification, determination of the location, and classification of damaged buildings are essential after destructive earthquakes. However, the accuracy of image change detection is limited because of the many texture features and changes in non-building information. In this context, a model for single-building damage detection based on multi-feature fusion is proposed. First, the normalized Digital Surface Model (nDSM) was extracted from the DSM through iterative filtering and point cloud thinning, followed by the extraction of building contour information. Next, single-building images were generated from different data sources through the region of interest (ROI), and the optimal texture feature parameters were extracted for fusion. Afterward, principal-component analysis (PCA) was conducted to suppress multi-feature correlation-induced information redundancy. Finally, the damage to buildings was quantitatively evaluated, and the model was compared with 13 models. The results confirmed the practicability of the model for the Yangbi MS6.4 and Honghe MS5.0 earthquakes.

Spaceborne SAR Time-Series Images Change Detection Based on SAR-SIFT-Logarithm Background Subtraction

Synthetic Aperture Radar (SAR) image change detection aims to detect changes with images of the same area acquired at different times. It has wide applications in environmental monitoring, urban planning and resource management. Traditional change detection methods for spaceborne SAR time-series images typically adopt a pairwise comparison strategy to obtain multi-temporal change information. However, this kind of method has the problem of losing the overall change information, which is time consuming. To address this problem, this paper proposes a new change detection algorithm for spaceborne SAR time-series data based on SAR-SIFT-Logarithm Background Subtraction. This algorithm combines the SAR-SIFT image registration technology with Logarithm Background Subtraction. The method first preprocesses the input time-series data with steps like noise reducing and radiometric calibration. Then, the images will be coregistered by the SAR-SIFT step to avoid mismatches-induced detection performance degradation. Next, the parts that remained unchanged throughout the time period are modeled with a median filter to obtain the static background. The change information is then obtained via the subtraction of background and CFAR detection and clustering. The proposed algorithm is validated using the Sentinel-1 GRD and PAZ-1 time-series dataset. Experimental results demonstrate that the proposed method effectively detects the overall change information and reduces processing time compared to traditional pairwise comparison methods.

Change Detection Needs Neighborhood Interaction in Transformer

Remote sensing image change detection (CD) is an essential technique for analyzing surface changes from co-registered images of different time periods. The main challenge in CD is to identify the alterations that the user intends to emphasize, while excluding pseudo-changes caused by external factors. Recent advancements in deep learning and image change detection have shown remarkable performance with ConvNet-based and Transformer-based techniques. However, ConvNet-based methods are limited by the local receptive fields of convolutional kernels that cannot effectively capture the change features in spatial–temporal information, while Transformer-based CD models need to be driven by a large amount of data due to the lack of inductive biases, and at the same time need to bear the costly computational complexity brought by self-attention. To address these challenges, we propose a Transformer-based Siamese network structure called BTNIFormer. It incorporates a sparse attention mechanism called Dilated Neighborhood Attention (DiNA), which localizes the attention range of each pixel to its neighboring context. Extensive experiments conducted on two publicly available datasets demonstrate the benefits of our proposed innovation. Compared to the most competitive recent Transformer-based approaches, our method achieves a significant 12.00% improvement in IoU while reducing computational costs by half. This provides a promising solution for further development of the Transformer structure in CD tasks.

A Full-Scale Connected CNN–Transformer Network for Remote Sensing Image Change Detection

Recent studies have introduced transformer modules into convolutional neural networks (CNNs) to solve the inherent limitations of CNNs in global modeling and have achieved impressive performance. However, some challenges have yet to be addressed: first, networks with simple connections between the CNN and transformer perform poorly in small change areas; second, networks that only use transformer structures are prone to attaining coarse detection results and excessively generalizing feature boundaries. In addition, the methods of fusing the CNN and transformer have the issue of a unilateral flow of feature information and inter-scale communication, leading to a loss of change information across different scales. To mitigate these problems, this study proposes a full-scale connected CNN–Transformer network, which incorporates the Siamese structure, Unet3+, and transformer structure, used for change detection in remote sensing images, namely SUT. A progressive attention module (PAM) is adopted in SUT to deeply integrate the features extracted from both the CNN and the transformer, resulting in improved global modeling, small target detection capacities, and clearer feature boundaries. Furthermore, SUT adopts a full-scale skip connection to realize multi-directional information flow from the encoder to decoder, enhancing the ability to extract multi-scale features. Experimental results demonstrate that the method we designed performs best on the CDD, LEVIR-CD, and WHU-CD datasets with its concise structure. In particular, based on the WHU-CD dataset, SUT upgrades the F1-score by more than 4% and the intersection over union (IOU) by more than 7% compared with the second-best method.

SAR Change Detection Algorithm Combined with FFDNet Spatial Denoising

Objectives: When detecting changes in synthetic aperture radar (SAR) images, the quality of the difference map has an important impact on the detection results, and the speckle noise in the image interferes with the extraction of change information. In order to improve the detection accuracy of SAR image change detection and improve the quality of the difference map, this paper proposes a method that combines the popular deep neural network with the clustering algorithm.Methods: Firstly, the SAR image with speckle noise was constructed, and the FFDNet architecture was used to retrain the SAR image, and the network parameters with better effect on speckle noise suppression were obtained. Then the log ratio operator is generated by using the reconstructed image output from the network. Finally, K-means and FCM clustering algorithms are used to analyze the difference images, and the binary map of change detection results is generated. Results: The experimental results have high detection accuracy on Bern and Sulzberger's real data, which proves the effectiveness of the method.

Unsupervised change detection in PolSAR images using siamese encoder–decoder framework based on graph-context attention network

Unsupervised change detection in PolSAR images using siamese encoder–decoder framework based on graph-context attention network

A dual-branch siamese spatial-spectral transformer attention network for Hyperspectral Image Change Detection

The convolutional neural networks have recently gained widespread attention in Hyperspectral Image Change Detection (HSI-CD) due to their outstanding feature extraction ability. However, limited by the inherent network backbones, the convolutional neural networks (CNNs) fail to mine the sequence attributes and model the intricate relationships of spectral signatures. In contrast, transformers are proficient at learning sequence information owing to the powerful self-attention mechanisms. The two backbone structures exhibit complementary spatial and spectral feature extraction strengths, respectively. Inspired by this, we propose a dual-branch siamese spatial–spectral transformer attention network (DBS3TAN) for HSI-CD. The main idea is to fully exploit the advantages of CNNs and transformers for spatial and spectral feature extraction. More importantly, we devise the two key modules, i.e., the spatial attention module and the spatial–spectral transformer module. The former utilizes depthwise separable convolutions and attention mechanisms to emphasize the features of dual-temporal HSIs from the spatial perspective. The latter focuses on the sequence attributes of spectral signatures and mines the spatial characteristics from adjacent pixels. We employ the weighted contrastive loss function to separate the changed and unchanged pixels more reliably and set the random weight factors to balance the contributions of the two branches. Finally, the threshold values judgment is used to obtain the ultimate detection maps. We conduct extensive experiments to evaluate the DBS3TAN on three HSI datasets, demonstrating its superior performances than compared methods qualitatively and quantitatively. The source code will be available at https://github.com/zhangyiyan001/DBS3TAN.

Assessing the impact of villagization program on land use land cover dynamics in Benishangul-Gumuz, Western Ethiopia

<p>Planning for continuing natural resource management requires current information on the dynamics of land use and land cover. The aim of this paper was to analyze the impacts of the villagization program on land use land cover dynamics in<strong> </strong>Benishangul-Gumuz region, western Ethiopia. The study has employed a mixed-method research design using both primary and secondary sources. Multispectral LANDSAT satellite images with a 30 m resolution were acquired for land use land cover change detection between the years 1999, 2009, and 2022. Arc GIS 10.8, QGIS 3.28, ERDAS Imagine 2014, and Microsoft Excel software were used for image classification, accuracy assessment, and change detection. Six different land use land cover types: forest land, shrub and grassland, cultivated land, residential, bare land, and water bodies were identified between 1999 and 2022. The trends indicated a dramatic decrease at the rate of 27.2 ha of forestland, 17.1 ha of shrub and grassland, and 4.6 ha of water bodies per year, while the share of cultivated land, residential, and bare land has expanded at an average rate of 34.3 ha, 11.7 ha, and 2.9 ha per year respectively between 1999 and 2022. The phenomenon was caused by added population pressure due to villagization program, which in turn triggered farmland expansion and deforestation. It is recommended that raising local community awareness, reforestation, practicing land use plans, and promoting successful livelihood diversification could help to alleviate the issue and reroute the course of events in order to achieve sustainable natural resource management.</p>

Temporal difference-guided network for hyperspectral image change detection

ABSTRACT Recently, the research area of hyperspectral (HS) image change detection (CD) is popular with convolutional neural networks (CNNs) based methods. However, conventional CNNs-based CD algorithms commonly achieve detection by comparing the deep features extracted from the bi-temporal images at decision level, which often fails to take full advantage of the features extracted by the network at different levels. Moreover, there are inevitably substantial redundant features located in non-varying regions in bi-temporal images, which considerably impedes the training efficiency of CNNs-based methods. To solve these two problems, we propose a temporal difference-guided HS image CD network, called TDGN Specifically, the rich spectral features will be extracted from the bi-temporal images hierarchically, and then the differences between the two images at different levels of the network will be yielded by the elaborated convolutional gated recurrent unit block in the spatial dimension. Furthermore, the differences from these different levels will be fused for the final detection. More significantly, to boost the efficiency of the backbone network for feature extraction, the obtained difference at each level is also leveraged to generate variation weights to guide the feature extraction at the next stage. Finally, the proposed TDGN can make full use of the temporal difference obtained by the network at different levels while this information is further employed to facilitate the attention and extraction of change features by the network. Extensive experiments, implemented on four well-known HS data sets, demonstrate that the proposed TDGN yields an average overall accuracy of 98.67%, 96.74%, 99.36%, and 96.81% on these data sets, respectively, acquiring promising detection performance compared to state-of-the-art methods. The codes of this work will be available at https://github.com/zhonghaochen/TDGN_Master for the sake of reproducibility.

CSANet: a channel-spatial attention network for remote sensing image change detection

ABSTRACT Deep learning has revolutionized change detection (CD) in remote sensing tasks. However, high-resolution CD based on deep learning faces challenges in handling semantic complexity in bi-temporal images, especially across varying weather and lighting conditions. While CNN-based approaches optimize network structures to enhance contextual information, recent attention-based methods increase computational demands. This study introduces the Channel-Spatial Attention Network (CSANet) to extract multi-scale and semantic information from images. Evaluated on LEVIR-CD and DSIFN-CD datasets, CSANet outperforms several state-of-the-art methods, demonstrating its potential for advanced change detection in high-resolution remote sensing.