Search

Integration of GIS and remote sensing is a powerful source for implementing temporal analysis. The focus of this paper was to use Landsat 5 and 8 imageries to monitor changes in the Ranya city throughout five time periods (2000, 2005, 2010, 2015, and 2020). Image classification was done using iso cluster unsupervised classification tool via ArcGIS. To calculate the kappa coefficient and assess the accuracy, random points of the outcome were compared with ground truth data using Google Earth Pro, which produced highly accurate results. According to this study, the Ranya city had the highest proportion of green space (52%) in 2000. Though, after 20 years due to rapid urbanization, Ranya's built-up regions had quickly outgrown both bare land and agricultural land, growing in size by 223.53%. However, from the year 2000 to 2020, green spaces and bare land both showed a clear downward trend, losing, respectively, 44.9% and 41.3% of their space. Being an independent administration is one of the key causes of Ranya city's rapid urbanization and subsequent development between 2000 and 2020, in addition to other aspects like the building of numerous government and residential facilities.

Abstract Wildfires are uncontrolled fires fuelled by dry conditions, high winds, and flammable materials that profoundly impact vegetation, leading to significant consequences including noteworthy changes to ecosystems. In this study, we provide a novel methodology to understand and evaluate post-fire effects on vegetation. In regions affected by wildfires, earth-observation data from various satellite sources can be vital in monitoring vegetation and assessing its impact. These effects can be understood by detecting vegetation change over the years using a novel unsupervised method termed Deep Embedded Clustering (DEC), which enables us to classify regions based on whether there has been a change in vegetation after the fire. Our model achieves an impressive accuracy of 96.17%. Appropriate vegetation indices can be used to evaluate the evolution of vegetation patterns over the years; for this study, we utilized Enhanced Vegetation Index (EVI) based trend analysis showing the greening fraction, which ranges from 0.1 to 22.4 km2 while the browning fraction ranges from 0.1 to 18.1 km2 over the years. Vegetation recovery maps can be created to assess re-vegetation in regions affected by the fire, which is performed via a deep learning-based unsupervised method, Adaptive Generative Adversarial Neural Network Model (AdaptiGAN) on post-fire data collected from various regions affected by wildfire with a training error of 0.075 proving its capability. Based on the results obtained from the study, our approach tends to have notable merits when compared to pre-existing works.

Abstract The change detection (CD) technology has greatly improved the ability to interpret land surface changes. Deep learning (DL) methods have been widely used in the field of CD due to its high detection accuracy and application range. DL-based CD methods usually cannot fuse the extracted feature information at full scale, leaving out effective information, and commonly use transfer learning methods, which rely on the original dataset and training weights. To address the above issues, we propose a deeply supervised (DS) change detection network (DASUNet) that fuses full-scale features, which adopts a Siamese architecture, fuses full-scale feature information, and realizes end-to-end training. In order to obtain higher feature information, the network uses atrous spatial pyramid pooling (ASPP) module in the coding stage. In addition, the DS module is used in the decoding stage to exploit feature information at each scale in the final prediction. The experimental comparison shows that the proposed network has the current state-of-the-art performance on the CDD and the WHU-CD, reaching 94.32% and 90.37% on F1, respectively.

Abstract Land use and land cover (LULC) analysis is crucial for understanding societal development and assessing changes during the Anthropocene era. Conventional LULC mapping faces challenges in capturing changes under cloud cover and limited ground truth data. To enhance the accuracy and comprehensiveness of the descriptions of LULC changes, this investigation employed a combination of advanced techniques. Specifically, multitemporal 30 m resolution Landsat-8 satellite imagery was utilized, in addition to the cloud computing capabilities of the Google Earth Engine (GEE) platform. Additionally, the study incorporated the random forest (RF) algorithm. This study aimed to generate continuous LULC maps for 2014 and 2020 for the Shrirampur area of Maharashtra, India. A novel multiple composite RF approach based on LULC classification was utilized to generate the final LULC classification maps utilizing the RF-50 and RF-100 tree models. Both RF models utilized seven input bands (B1 to B7) as the dataset for LULC classification. By incorporating these bands, the models were able to influence the spectral information captured by each band to classify the LULC categories accurately. The inclusion of multiple bands enhanced the discrimination capabilities of the classifiers, increasing the comprehensiveness of the assessment of the LULC classes. The analysis indicated that RF-100 exhibited higher training and validation/testing accuracy for 2014 and 2020 (0.99 and 0.79/0.80, respectively). The study further revealed that agricultural land, built-up land, and water bodies have changed adequately and have undergone substantial variation among the LULC classes in the study area. Overall, this research provides novel insights into the application of machine learning (ML) models for LULC mapping and emphasizes the importance of selecting the optimal tree combination for enhancing the accuracy and reliability of LULC maps based on the GEE and different RF tree models. The present investigation further enabled the interpretation of pixel-level LULC interactions while improving image classification accuracy and suggested the best models for the classification of LULC maps through the identification of changes in LULC classes.

The agricultural sector's output holds paramount significance for the global population, serving as an indispensable resource for survival and consumption. Consequently, alterations in agricultural landscapes bear substantial implications for the world's food supply. The objectives of this research are to investigate the depletion of agricultural land, with a specific focus on Samut Songkhram Province—an agriculturally prominent region in Thailand renowned for supplying seafood and fruits to Bangkok. By employing advanced remote sensing and change detection methods and incorporating indices like NDVI, NDWI, and NDBI, the study meticulously analyzed land-use changes. The outcomes were rigorously scrutinized through supervised classification, validated by on-site inspections, and corroborated with data from pertinent agencies. Findings revealed that Samut Songkhram had sustained its prominence in agricultural land, constituting around 70% of the province's total area over the past two decades. However, this expanse has undergone persistent transformation during the last 20 years. Notably, the most substantial surge was observed in the conversion of agricultural land to urban and developed areas, particularly in the urban zones of Amphawa District, followed by Mueang Samut Songkhram and Bang Khonthi districts. This investigation illuminates a consistent downward trend in agricultural land, a vital source of sustenance for Thailand's population and the global community. Doi: 10.28991/ESJ-2024-08-02-020 Full Text: PDF

Remote sensing image change detection aims to identify differences between images acquired at different times in the same area, crucial for land management, environmental monitoring, and disaster assessment. Current change detection methods mainly use Siamese or early fusion structures. Siamese networks focus on object features at different times but lack attention to change information, leading to false alarms and missed detections. Early fusion structures focus on fused features but neglect temporal object features, hindering accurate change detection. To address these issues, we propose a novel network, Triplet UNet (T-UNet), employing a triplet encoder to simultaneously extract object features and change features between the pre- and post-time-phase images. To effectively interact features extracted from the three branches of triplet encoder, we propose a multi-branch spatial-spectral cross-attention module. In the decoder, we employ channel attention and spatial attention mechanisms to fully mine and integrate detailed texture and semantic localization information. The proposed T-UNet surpasses seven other state-of-the-art methods on three publicly available datasets. Extensive experiments verify the effectiveness of the proposed structure and modules as well as the superiority of the proposed T-UNet. The source code for the proposed T-UNet is accessible at https://github.com/Pl-2000/T-UNet.

ABSTRACTChange in land use land cover (LULC) is driven by human activities and drives changes that limit the accessibility of services and products for livestock and humans. This study aimed to develop the spatio-temporal changes in LULC, vegetation cover, and moisture index using multi-temporal satellite data in Southern Punjab, Pakistan. Analysis of satellite data was accomplished using ArcGIS 10.4 software. Based on ground-truthing, the supervised classification technique (maximum likelihood algorithm) was used to achieve the LULC classification. The LULC change analysis revealed that vegetation area converted to the build-up area by 7.17 % and bare soil converted to urban areas by 1.68 % from 2000 to 2021. Average NDVI values were calculated at 0.23, 0.17, 0.19, and 0.14 for 2000, 2007, 2014, and 2021, respectively. In the study area, average NDMI values were observed at 0.28, 0.25, 0.2, and 0.15 for 2000, 2007, 2014, and 2021, respectively. Based on our study, the general trend in Southern Punjab is a decrease in vegetation cover, moisture index, and forest area due to an increase in build-up areas. Assessments of LULC and NDMI changes, as well as estimates of the effect on the environment, are necessary for many policy decisions and planning.

Interactions between tropical cyclones and wildfires occur widely and can tip closed forests into open-canopy structures that initiate a 'grass–fire' cycle. We examined cyclone–fire interactions in a monodominant tropical montane pine forest in the Dominican Republic using remotely-sensed imagery to quantify damage from fires between 1986 and 2004, a category 1 cyclone in 1998, and an extensive wildfire in 2005. We also measured forest structure and composition 14.7 years after the 2005 fire. The area inside the 2005 burn scars (fire perimeters) totaled 25,206 ha, of which 81% burned and 14% was cyclone damaged. Cyclone damage made the fire markedly more extensive and severe—high-severity fires were > 3 times more frequent with high-severity cyclone damage than no cyclone damage—but these markedly synergistic effects were restricted to areas that had not burned for at least 19 years before the 2005 fire. Though earlier fires from 1986 to 2004 were small and low-severity, they were sufficient, when present, to prevent high-severity fire in 2005 irrespective of cyclone severity. In areas with strong cyclone–fire interactions, there was a complete loss of pine canopies, yet these stands had abundant pine canopy recruitment by 2019 and showed no evidence of compositional shifts toward open-canopy structures with pyrogenic herbaceous understories, illustrating the resilience of this ecosystem to a range of cyclone–fire synergies. However, the future resilience of tropical montane pine forests to cyclone–fire synergies is uncertain as climate change increases the intensity of cyclones and frequency of drought-triggered fires in these ecosystems. [ABSTRACT FROM AUTHOR]

Research on transformers in remote sensing (RS), which started to increase after 2021, is facing the problem of a relative lack of review. To understand the trends of transformers in RS, we undertook a quantitative analysis of the major research on transformers over the past two years by dividing the application of transformers into eight domains: land use/land cover (LULC) classification, segmentation, fusion, change detection, object detection, object recognition, registration, and others. Quantitative results show that transformers achieve a higher accuracy in LULC classification and fusion, with more stable performance in segmentation and object detection. Combining the analysis results on LULC classification and segmentation, we have found that transformers need more parameters than convolutional neural networks (CNNs). Additionally, further research is also needed regarding inference speed to improve transformers' performance. It was determined that the most common application scenes for transformers in our database are urban, farmland, and water bodies. We also found that transformers are employed in the natural sciences such as agriculture and environmental protection rather than the humanities or economics. Finally, this work summarizes the analysis results of transformers in remote sensing obtained during the research process and provides a perspective on future directions of development. [ABSTRACT FROM AUTHOR]

Wildfires are uncontrolled fires fuelled by dry conditions, high winds, and flammable materials that profoundly impact vegetation, leading to significant consequences including noteworthy changes to ecosystems. In this study, we provide a novel methodology to understand and evaluate post-fire effects on vegetation. In regions affected by wildfires, earth-observation data from various satellite sources can be vital in monitoring vegetation and assessing its impact. These effects can be understood by detecting vegetation change over the years using a novel unsupervised method termed Deep Embedded Clustering (DEC), which enables us to classify regions based on whether there has been a change in vegetation after the fire. Our model achieves an impressive accuracy of 96.17%. Appropriate vegetation indices can be used to evaluate the evolution of vegetation patterns over the years; for this study, we utilized Enhanced Vegetation Index (EVI) based trend analysis showing the greening fraction, which ranges from 0.1 to 22.4 km2 while the browning fraction ranges from 0.1 to 18.1 km2 over the years. Vegetation recovery maps can be created to assess re-vegetation in regions affected by the fire, which is performed via a deep learning-based unsupervised method, Adaptive Generative Adversarial Neural Network Model (AdaptiGAN) on post-fire data collected from various regions affected by wildfire with a training error of 0.075 proving its capability. Based on the results obtained from the study, our approach tends to have notable merits when compared to pre-existing works. [ABSTRACT FROM AUTHOR]

Heterogeneous image change detection is a very practical and challenging task because the data in the original image have a large distribution difference and the labeled samples of the remote sensing image are usually very few. In this study, we focus on solving the issue of comparing heterogeneous images without supervision. This paper first designs a self-paced multi-scale joint feature mapper (SMJFM) for the mapping of heterogeneous data to similar feature spaces for comparison and incorporates a self-paced learning strategy to weaken the mapper's capture of non-consistent information. Then, the difference information in the output of the mapper is evaluated from two perspectives, namely noise robustness and detail preservation effectiveness; then, the change detection problem is modeled as a multi-objective optimization problem. We decompose this multi-objective optimization problem into several scalar optimization subproblems with different weights, and use particle swarm optimization to optimize these subproblems. Finally, the robust evaluation strategy is used to fuse the multi-scale change information to obtain a high-precision binary change map. Compared with previous methods, the proposed SMJFM framework has the following three main advantages: First, the unsupervised design alleviates the dilemma of few labels in remote sensing images. Secondly, the introduction of self-paced learning enhances SMJFM's capture of the unchanged region mapping relationship between heterogeneous images. Finally, the multi-scale change information fusion strategy enhances the robustness of the framework to outliers in the original data. [ABSTRACT FROM AUTHOR]

The necessity for precise and current data concerning the dynamics of land cover change in Indonesia is crucial for efforts to reduce natural vegetation cover due to agricultural expansion. The functionality of monitoring systems that incorporate Terra-MODIS is currently compromised by the limited availability of data for the immediate future. This study seeks to assess the potential of VIIRS satellite imagery in developing an early warning system for monitoring vegetation cover change in Indonesia. The normalized differential open-area index (NDOAI) computed from 8-day VIIRS data was employed to detect changes in vegetation cover based on pixel-by-pixel subtraction in the NDOAI data time series. Evaluating the pixel-level accuracy of change detection is complicated due to the fact that we evaluate a change map at a coarser resolution than the Landsat-based reference map. The results revealed that increasing the threshold percentage is associated with improved accuracy. In change detection, there is often a trade-off between accuracy and sensitivity. A threshold that is too low may result in false positives, while a threshold that is too high may lead to missed changes. This study demonstrates that when a threshold value of less than 20% is applied, Landsat can identify vegetation cover changes at an earlier stage. Conversely, when a threshold value greater than 20% is employed, the VIIRS will detect the change 4.5 days earlier than Landsat. Additionally, the VIIRS is capable of detecting changes 25.4 days and 54.8 days faster than Landsat, respectively, when using thresholds of 40% and 70%. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) change detection provides a powerful tool for continuous, reliable, and objective observation of the Earth, supporting a wide range of applications that require regular monitoring and assessment of changes in the natural and built environment. In this paper, we introduce a novel SAR image change detection method based on principal component analysis and two-level clustering. First, two difference images of the log-ratio and mean-ratio operators are computed, then the principal component analysis fusion model is used to fuse the two difference images, and a new difference image is generated. To incorporate contextual information during the feature extraction phase, Gabor wavelets are used to obtain the representation of the difference image across multiple scales and orientations. The maximum magnitude across all orientations at each scale is then concatenated to form the Gabor feature vector. Following this, a cascading clustering algorithm is developed within this discriminative feature space by merging the first-level fuzzy c-means clustering with the second-level neighbor rule. Ultimately, the two-level combination of the changed and unchanged results produces the final change map. Five SAR datasets are used for the experiment, and the results show that our algorithm has significant advantages in SAR change detection. [ABSTRACT FROM AUTHOR]

The challenge of detecting changes in high-resolution remote sensing imagery often stems from the difficulties in effectively extracting features and constructing appropriate change detection models considering the scale characteristics of ground objects. To solve these issues, we propose a novel UNet 3+ change detection method that considers the scale characteristics inherent in various land-cover change types. Our method includes three key steps: a multi-scale segmentation method, a class-specific UNet 3+ method, and an object-oriented change detection method based on UNet 3+. To verify the effectiveness of this method, we select two datasets for experiments and compare our proposed method with the UNet 3+ single-scale sampling method, the class-specific UNet 3+ single-scale sampling method, and the UNet 3+ multi-scale hierarchical sampling method. Our experimental results show that our proposed method has higher overall accuracy and F1, lower missed detection rate and false detection rate, and can detect more changes in ground features than other methods. To verify the scalability of this method, we compare this method with traditional change detection methods such as PCA-k-means, OCVA, a single-scale sampling method based on random forest, and a class-specific object-based method. Experimental results and accuracy indexes show that our proposed method better considers the scale characteristics of ground objects and achieves higher accuracy. Additionally, we compared our proposed method with other DLCD methods including LamboiseNet, BIT, CDNet, FCSiamConc, and FCSiamDiff. Our results show that our proposed method effectively considers edge information and has an acceptable time consumption. Our approach not only considers the full-scale characteristics of the feature extraction but also the scale characteristics of the change detection model. In addition, it considers a more practical feature extraction unit (object), making it more accurate. [ABSTRACT FROM AUTHOR]

Landslide event inventories are one of the most critical datasets to increase knowledge on landslide occurrences. However, they are rarely available in various regions, especially in countries of the Global South. This study aims to generate rainfall-induced landslide event inventories and define the rainfall thresholds responsible for landslide occurrence at the national scale of Malawi, Africa. We mainly followed a three-step methodology to generate landslide inventories. First, we went through media reports to identify documented landslide events. Second, we used Sentinel-2 images to identify possible areas affected by landslides using automated change detection algorithms based on vegetation indices. Third, we manually went through optical images provided by Planet Lab and Google Earth and mapped landslides via visual image interpretation. Overall, we mapped 27 rainfall-induced landslide inventories between 2003 and 2022, with a total of 4709 individual landslides. We then analysed the Malawian terrain and identified two different landscape clusters (i.e. Cluster 1 and Cluster 2) showing similar morphometric and climatic conditions. Ultimately, we calculated the rainfall threshold for each landscape cluster. The minimum rainfall amounts responsible for landsliding correspond to 66 mm/two-day and 51 mm/day in Clusters 1 and 2, respectively. In this context, our paper not only presents and shares the first national-scale, digital rainfall-induced landslide event inventory database of Malawi but also suitable rainfall thresholds to be potentially exploited for a national scale landslide early warning system. A similar framework could be applied to generate landslide inventories for other data scarce regions. [ABSTRACT FROM AUTHOR]

Al- Dalmaj area is a large water body in Iraq in the Afak district, which feeds from the general downstream between the Tigris and the Euphrates. The present study focused mainly on the changes detection in Al-Dalmaj area through analysis of the Landsat images for four selected years (2000, 2002, 2013 and 2017) to determine the changes in marsh waters, wetlands area, drylands areas and other land classes, The Landsat images of Al-Dalmaj area were analysed with the scene that includes the water surface and the area surrounded by the fixed geographical boundaries. The combination of remote sensing and GIS techniques have been found to be successful and effective in monitoring the spatial and temporal changes in land cover. These changes in the land cover of Al-Dalmaj area were detected using various analyses, including Normalised Difference Vegetation (NDVI), the Normalised Difference Water Index (NDWI), the Salinity Index(SI), and the Salinity ratio using Geographic Information Systems (GIS) programs v.10, using ERDAS Imagine v. 9.2 and spectral index techniques were used in this study to determine the land changes, to analyse Landsat images for the years (2000,2002,2013 and 2017). From the application of the equations of evidence referred to, it was found that the salinity (%) increased in the study area. 2002 represents the most extreme stage of the lack of natural conditions in Al- Dalmaj area. It was concluded that the water change was 83.8% for 2002 as 2000 as a reference. As for the natural vegetation, the percentage of change was 84.5% for the same year. The results showed that the percentage change in salinity was 127.5%, a high percentage compared to the rest of the study years. The bulk density values range between (0.70 - 0.89) Mg/m³. [ABSTRACT FROM AUTHOR]

The task of change detection (CD) in optical and SAR images is an ever-evolving and demanding subject within the realm of remote sensing (RS). It holds great significance to identify the target areas by using complementary information between the two. Due to the distinct imaging mechanisms employed by optical and SAR sensors, effectively and accurately identifying changing regions can be challenging. To this end, a novel heterogeneous RS images CD network (Twin-Depthwise Separable Convolution Connect Network, TDSCCNet) is proposed in this paper. Image domain transformation is a front-end task, while the back-end employs a single-branch bilayer depthwise separable convolutionconnected encoder-decoder to accomplish CD work. Specifically, first, the cycle-consistent adversarial network (CycleGAN) serves to integrate the optical and SAR visual domains, and a consistent feature expression is obtained. Second, the single-branch encoder structure of bilayer depthwise separable convolution is employed to realize change feature extraction. Finally, the multiscale connected decoder reconstructed by change map is utilized to reconstruct the original images and solve the local discontinuities and holes in the binary change map. Multiscale loss is designated for optimizing the global and local effects to alleviate the class imbalance problem. It was tested on four representative datasets from Gloucester, Shuguang Village, Italy and WV-3 datasets with an overall accuracy of 97.36%, 97.01%, 97.62%, and 98.01% respectively. By comparing the existing methods, experimental results confirmed the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Change detection (CD) in remote sensing (RS) images aims to identify surface changes based on images acquired at different times. However, existing methods are still unsatisfactory in locating fine details of change in RS images, due to overlooking the inherent temporal information. To address the issue, we introduce a novel Triplet Fusion Temporal Relationship Network (FTRNet). FTRNet incorporates a triplet input backbone that enables the extraction of both spatial and temporal features. We design a change attention module to enhance bitemporal features, making the backbone network retain temporal information and fuse cross-scale features to extract the high-level location information. We evaluate our method on three benchmark datasets, including LEVIR-CD, WHU-CD, GZ, and DSIFN. The experimental results showcase that FTRNet achieves IoU scores of 83.60%, 77.06%, 73.67%, and 77.30% in LEVIR-CD, WHU-CD, GZ, and DSIFN datasets, respectively. These results surpass the second-best baseline by 1.20%, 0.49%, 1.31%, and 1.20%, respectively. [ABSTRACT FROM AUTHOR]

Cropland non-agriculturalization (CNA) refers to the conversion of cropland into construction land, woodland/garden/grassland, water body, or other non-agricultural land, which ultimately disrupts local agroecosystems and the cultivation and production of crops. Remote sensing technology is an important tool for large-area CNA detection, and remote sensing based methods that can be used for this task include the time-series analysis method and change detection from bi-temporal images. In particular, change detection methods using high-resolution remote sensing imagery have great potential for CNA detection, but enormous challenges do still remain. The large intra-class variance of cropland with different phenological stages and planting patterns leads to cropland areas being difficult to identify effectively, while certain features can be misidentified because they are similar to cropland, resulting in false alarms and missed detections in the results. There is also a lack of large-scale CNA datasets covering multiple change scenarios as data support. To address these problems, a lightweight model focused on CNA detection (CNANet) is proposed in this paper. Specifically, the uniquely crafted represent-consist-enhance (RCE) module is seamlessly integrated between the encoder and decoder components of CNANet to perform a contrast operation on the deep features extracted by the feature extractor. The RCE module is specifically designed to aggregate multiple cropland representations and extend the cropland representations from the confusing background, to achieve the purpose of reducing the intra-class reflectance differences and enhancing the model's perception of cropland. In addition, a large-scale high-resolution cropland non-agriculturalization (Hi-CNA) dataset was built for the CNA identification task, with a total of 6797 pairs of 512 × 512 images with semantic annotations. Compared to the existing datasets, the Hi-CNA dataset has the advantages of multiple phenological stages, multiple change scenarios, and multiple annotation types, in addition to the large data volume. The experimental results obtained in this study show that the benchmark methods tested on the Hi-CNA dataset can all achieve a good accuracy, proving the high-quality annotation of the dataset. The overall accuracy and F1-score of CNANet with the default settings reach 93.81 % and 78.9 %, respectively, achieving a superior accuracy, compared to the other benchmark methods, and demonstrating stronger perception of cropland changes. In addition, in two selected verification regions within the large-scale real-world CNA mapping results, the F1-score is 83.61 % and 50.87 %. The Hi-CNA can be downloaded from http://rsidea.whu.edu.cn/Hi-CNA_dataset.htm. [ABSTRACT FROM AUTHOR]

The existing change detection methods are difficult to balance local features and global features, and the boundary between change objects and backgrounds is blurred, so this paper proposed a remote sensing image building change detection method based on local-global feature coupling and boundary guidance. In the encoding stage, the method adopted parallel convolutional neural network and Transformer to extract the local features and global representation of remote sensing images, respectively. At different scales, the local-global feature coupling module fused local features and global feature representation to enhance the expression ability of image features. In addition, it introduced the boundary guidance branch to obtain the prior boundary information of the change objects, so that its guide change map can highlight the structural characteristics of the building and promote the accurate boundary location. This paper conducted experiments on the LEVIR-CD and WHU datasets, resulting in F1-score of 91.25% and 91.27%, IoU of 83.90% and 83.95%, respectively. The experimental results show that the method has a great improvement in the detection accuracy and good generalization ability. [ABSTRACT FROM AUTHOR]

This letter proposes a novel multi-task contrastive learning (MTCL) approach for change detection of high-resolution remote sensing images. Current self-supervised learning approaches have limitations in utilizing multiview information of remote sensing images, and they cannot directly train the change detection backbone network for feature learning. The investigated method exploits multiview information contained in remote sensing images to directly train the change detection backbone network. Specifically, multiview views are constructed from high-resolution remote sensing images by handcrafted feature extraction approach, which contain complementary information from different domains. Then each constructed view is utilized for contrastive learning, and the multi-task learning strategy of parameter sharing is used to learn more rich and robust information. Furthermore, instead of learning feature representations from single remote sensing image, the proposed approach directly utilizes image pairs to train the change detection backbone, further improving the performance of change detection network. Two benchmark datasets are employed for comparative experiments, and experimental results certify the effectiveness and superiority of the investigated approach. [ABSTRACT FROM AUTHOR]