Search

Change detection in high-resolution Earth observation is a fundamental Earth vision task to understand the subtle temporal dynamics of Earth's surface, significantly promoted by generic vision technologies in recent years. Vision Transformer is a powerful component to learning spatiotemporal representation but with enormous computation complexity, especially for high-resolution images. Besides, there is still lacking principles in designing macro architectures integrating these advanced vision components for various change detection tasks. In this paper, we present a deep probabilistic change model (DPCM) to provide a unified, interpretable, modular probabilistic change process modeling to address multiple change detection tasks, including binary change detection, one-to-many semantic change detection, and many-to-many semantic change detection. DPCM describes any complex change process as a probabilistic graphical model to provide theoretical evidence for macro architecture design and generic change detection task modeling. We refer to this probabilistic graphical model as the probabilistic change model (PCM), where DPCM is the PCM parameterized by deep neural networks. For parameterization, the PCM is factorized into many easy-to-solve distributions based on task-specific assumptions, and then we can use deep neural modules to parameterize these distributions to solve the change detection problem uniformly. In this way, DPCM has both theoretical macro architecture from PCM and strong representation capability of deep networks. We also present the sparse change Transformer for better parameterization. Inspired by domain knowledge, i.e., the sparsity of change and the local correlation of change, the sparse change Transformer computes self-attention within change regions to model spatiotemporal correlations, which has a quadratic computational complexity of the change region size but independent of image size, significantly reducing computation overhead for high-resolution image change detection. We refer to this instance of DPCM with sparse change Transformer as ChangeSparse to demonstrate their effectiveness. The experiments confirm ChangeSparse's superiority in speed and accuracy for multiple real-world application scenarios, such as disaster response and urban development monitoring. The code is available at https://github.com/Z-Zheng/pytorch-change-models. More resources can be found in http://rsidea.whu.edu.cn/resource_sharing.htm. [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 natural terrain (e.g., farm and forest) in temperate zones changes dramatically between seasons due to distinct temperatures and precipitation variations from summer to winter. Moreover, built-up areas vary little in this short period. Therefore, extracting built-up areas via change detection on polarimetric synthetic aperture radar (PolSAR) images is feasible. A common type of PolSAR change detection method is based on hypothesis testing theory. However, in these methods, pixels are selected as the processing units; as a result, these models are computationally complex and poorly maintain the boundaries of built-up areas. In this paper, we innovatively introduce superpixels into the hypothesis test theory and propose a superpixelwise PolSAR change detection method for built-up area extraction. First, we oversegment the PolSAR images into a set of superpixels and derive the probability density function (PDF) of a superpixel's reflectivity on a PolSAR image. Based on this distribution, we present a superpixelwise likelihood-ratio test (LRT) statistic to measure the similarity of two superpixelwise covariance matrices for unsupervised change detection. When actually computing the superpixelwise LRT, the large variation in the areas of the superpixels makes the likelihood functions very complex and estimating the parameters difficult. We further simplify the calculation of the LRT statistic and apply it to built-up area extraction. Compared to the state-of-the-art built-up area extraction methods, our approach provides the best results with overall accuracy values of 91.41%, 92.71%, 93.67% and 93.91% for the four studied areas respectively. In addition, the computational complexity of our method is assessed, and the run time (seconds) of the proposed method in the four study cases is 11.14, 6.32, 5.31, and 4.88, respectively, superior to the values of 66.12, 36.26, 33.65, and 29.67 for DRT, respectively. Code and datasets: https://github.com/SunXJ7/Sentinel-1-Datasets-for-Built-up-Area-Extraction. [ABSTRACT FROM AUTHOR]

Ghana's coastline has been facing erosion and sedimentation phenomena for several decades, resulting in a serious threat to life and property considering that major urban settlements are located on the coast. In this region, there has been a lack of emphasis on comprehensive, large-scale investigations into coastal changes: prior research has predominantly centered on site-specific assessments. These studies have revealed alarming erosion rates, with reports indicating that nearly ten meters are lost annually. The use of high-resolution remotely sensed data can be a consistent support in regions where physical or economic obstacles interfere with collecting in situ information. In particular, the use of continuous all-weather SAR data may facilitate the evaluation of erosion and sedimentation phenomena in coastal areas. In this paper, we apply SAR data over a time period between 2017 and 2021. Sentinel-1 data are pre-processed using the Google Earth Engine platform, and a dedicated algorithm is then applied to identify and quantify erosion and sedimentation processes. Optical images are used as a reference for detecting the location of two areas where consistent sedimentation and erosion phenomena occurred in the considered four years. The results demonstrate that SAR backscattering variations over time offer a reliable method for monitoring coastal changes. This approach enables the identification of the type of phenomena occurring - sedimentation or erosion -, and allows for the quantification of their intensity and dimensions over time. The method can be worldwide applied once the appropriate thresholds are evaluated and help in predictive studies and environmental planning. [ABSTRACT FROM AUTHOR]

In a constant evolving world, change detection is of prime importance to keep updated maps. To better sense areas with complex geometry (urban areas in particular), considering 3D data appears to be an interesting alternative to classical 2D images. In this context, 3D point clouds (PCs), whether obtained through LiDAR or photogrammetric techniques, provide valuable information. While recent studies showed the considerable benefit of using deep learning-based methods to detect and characterize changes into raw 3D PCs, these studies rely on large annotated training data to obtain accurate results. The collection of these annotations are tricky and time-consuming. The availability of unsupervised or weakly supervised approaches is then of prime interest. In this paper, we propose an unsupervised method, called DeepCluster 3D Change Detection (DC3DCD), to detect and categorize multiclass changes at point level. We classify our approach in the unsupervised family given the fact that we extract in a completely unsupervised way a number of clusters associated with potential changes. Let us precise that in the end of the process, the user has only to assign a label to each of these clusters to derive the final change map. Our method builds upon the DeepCluster approach, originally designed for image classification, to handle complex raw 3D PCs and perform change segmentation task. An assessment of the method on both simulated and real public dataset is provided. The proposed method allows to outperform fully-supervised traditional machine learning algorithm and to be competitive with fully-supervised deep learning networks applied on rasterization of 3D PCs with a mean of IoU over classes of change of 57.06% and 66.69% for the simulated and the real datasets, respectively. The code is available at https://github.com/idegelis/torch-points3d-dc3dcd. • DC3DCD, unsupervised learning method for multiclass 3D point cloud change detection. • Revisiting the DeepCluster principle for change segmentation in raw 3D point clouds. • Results are better than classical supervised machine learning methods. • Results are equal to rasterization-based deep learning supervised networks. • In a binary-to-multiclass scenario, DC3DCD competes with fully supervised methods. [ABSTRACT FROM AUTHOR]

In the context of Earth observation, change detection boils down to comparing images acquired at different times by sensors of possibly different spatial and/or spectral resolutions or different modalities (e.g., optical or radar). Even when considering only optical images, this task has proven to be challenging as soon as the sensors differ by their spatial and/or spectral resolutions. This paper proposes a novel unsupervised change detection method dedicated to images acquired by such so-called heterogeneous optical sensors. It capitalizes on recent advances which formulate the change detection task into a robust fusion framework. Adopting this formulation, the work reported in this paper shows that any off-the-shelf network trained beforehand to fuse optical images of different spatial and/or spectral resolutions can be easily complemented with a network of the same architecture and embedded into an adversarial framework to perform change detection. A comparison with state-of-the-art change detection methods demonstrates the versatility and the effectiveness of the proposed approach. • Detecting changes between images acquired by heterogeneous sensors is formulated as a robust fusion problem. • The method is able to handle multi-band images with possibly different spatial and/or spectral resolutions. • The method is unsupervised, i.e., it does not need a train set composed of image pairs with associated binary change maps. • Any pretrained deep network designed to fuse heterogeneous images can be reused to perform change detection. • When compared to conventional approaches, the strategy appears to be more flexible while competing favorably. [ABSTRACT FROM AUTHOR]

Change detection on multimodal remote sensing images has become an increasingly interesting and challenging topic in the remote sensing community, which can play an essential role in time-sensitive applications, such as disaster response. However, the modal heterogeneity problem makes it difficult to compare the multimodal images directly. This paper proposes a Fourier domain structural relationship analysis framework for unsupervised multimodal change detection (FD-MCD), which exploits both modality-independent local and nonlocal structural relationships. Unlike most existing methods analyzing the structural relationship in the original domain of multimodal images, the three critical parts in the proposed framework are implemented on the (graph) Fourier domain. Firstly, a local frequency consistency metric calculated in the Fourier domain is proposed to determine the local structural difference. Then, the nonlocal structural relationship graphs are constructed for pre-change and post-change images. The two graphs are then transformed to the graph Fourier domain, and high-order vertex information is modeled for each vertex by graph spectral convolution, where the Chebyshev polynomial is applied as the transfer function to pass K-hop local neighborhood vertex information. The nonlocal structural difference map is obtained by comparing the filtered graph representations. Finally, an adaptive fusion method based on frequency-decoupling is designed to effectively fuse the local and nonlocal structural difference maps. Experiments conducted on five real datasets with different modality combinations and change events show the effectiveness of the proposed framework. [ABSTRACT FROM AUTHOR]

Historical maps depict past states of the Earth's surface and make it possible to trace the natural or anthropogenic evolution of geographic objects back through time. However, the state of the depicted reality is not the only source of change: maps of varying age can differ in terms of graphical design, and also in terms of storage conditions, physical ageing of pigments, and the scanning process for digitization. Consequently, a computer vision system learned from a specific (source) map series will often not generalize well to older or newer (target) maps, calling for domain adaptation. In the present paper we examine – to our knowledge for the first time – domain adaptation for segmenting historical maps. We argue that for geo-spatial data like maps, which are geo-localized by definition, the spatial co-occurrence of geographical objects provides a supervision signal for domain adaptation. Since only a subset of all mapped objects co-occur, and even those are not perfectly aligned due to both real topographic changes and variations in map generalization/production, they only provide weak supervision — still they can bring a substantial benefit over completely unsupervised domain adaptation methods. The core of our proposed method is a novel self-supervised co-occurrence network that detects co-occurring objects across maps (specifically, domains) with a novel loss function that allows for object changes and spatial misalignment. Experiments show that, for the task of segmenting hydrological objects such as rivers, lakes and wetlands, our system significantly outperforms two state-of-art baselines, even with limited supervision (e.g., 5%). The source code is publicly available at https://github.com/sian-wusidi/spatialcooccurrence. [ABSTRACT FROM AUTHOR]

This study is concerned with urban change detection and categorization in point clouds. In such situations, objects are mainly characterized by their vertical axis, and the use of native 3D data such as 3D Point Clouds (PCs) is, in general, preferred to rasterized versions because of significant loss of information implied by any rasterization process. Yet, for obvious practical reasons, most existing studies only focus on 2D images for change detection purpose. In this paper, we propose a method capable of performing change detection directly within 3D data. Despite recent deep learning developments in remote sensing, to the best of our knowledge there is no such method to tackle multi-class change segmentation that directly processes raw 3D PCs. Thereby, based on advances in deep learning for change detection in 2D images and for analysis of 3D point clouds, we propose a deep Siamese KPConv network that deals with raw 3D PCs to perform change detection and categorization in a single step. Experimental results are conducted on synthetic and real data of various kinds (LiDAR, multi-sensors). Tests performed on simulated low density LiDAR and multi-sensor datasets show that our proposed method can obtain up to 80% of mean of IoU over classes of changes, leading to an improvement ranging from 10% to 30% over the state-of-the-art. A similar range of improvements is attainable on real data. Then, we show that pre-training Siamese KPConv on simulated PCs allows us to greatly reduce (more than 3,000 ×) the annotations required on real data. This is a highly significant result to deal with practical scenarios. Finally, an adaptation of Siamese KPConv is realized to deal with change classification at PC scale. Our network overtakes the current state-of-the-art deep learning method by 23% and 15% of mean of IoU when assessed on synthetic and public Change3D datasets, respectively. The code is available at the following link: https://github.com/IdeGelis/torch-points3d-SiameseKPConv. • Siamese KPConv, the first deep network for change segmentation in 3D point clouds. • Urb3DCD–V2, a new simulated dataset to assess 3D point cloud change detection methods. • Pre-training Siamese KPConv on simulated data highly boosts the training on real data. [ABSTRACT FROM AUTHOR]

Terrestrial laser scanning (TLS) has gained prominence as a deformation monitoring technology in various fields such as bridges and tunnels. This paper offers a comprehensive overview of TLS-based deformation monitoring, covering measurement principles, stochastic models, registration, change detection, and practical applications. We systematically classify, detail, and compare methods for change detection, registration, and georeferencing, shedding light on their strengths, weaknesses, and applicability. The paper also provides an extensive comparison of deformation detection techniques for structures like bridges and tunnels, as well as ground deformation in landslides and unstable rock masses. Challenges related to achieving high-precision, reliable, and efficient deformation monitoring are discussed, and avenues for future research are outlined. Emphasis is placed on the importance of measurement planning to ensure data accuracy and support informed decision-making, particularly for large or spatially constrained structures. Future research should explore leveraging geometric constraints for more accurate change detection and enhancing deformation analysis reliability. • TLS-based deformation monitoring shows commonalities and specific characteristics across scenarios. • Multi-epoch point cloud comparison is essential for all types of deformation monitoring applications. • Measurement planning is crucial for accurate and reliable monitoring in complex structures. • Future research should leverage geometric constraints for precise analysis of deformation. • Research is needed to accurately quantify deformations in monitoring irregular ground surfaces. [ABSTRACT FROM AUTHOR]

• A novel deep learning model for laser-induced damage change detection is proposed. • The proposed model combines siamese CNN with visual attention mechanism. • A novel SoftmaxFocal loss is put forward to train the network. • Our model outperforms four different methods overall on three real datasets. With high-energy laser irradiating, the laser-induced damages may occur in the surfaces of optical elements in laser facilities. As the laser-induced damage changes can badly affect regular and healthy operation of laser facilities, it is essential to effectively detect real damage changes while suppressing meaningless and spurious changes in captured optical images. In order to achieve high-precision laser-induced damage change detection, this paper presents a novel deep learning model which exploits visual attention-based siamese convolutional neural network with SoftmaxFocal loss and significantly improves the performance of damage change detection. In the proposed model, an end-to-end classification network is designed and trained which fuses the spatial-channel domain collaborative attention modules into siamese convolutional neural network thus achieving more efficient feature extraction and representation. For the purpose of addressing the unbalanced distribution of hard and easy samples, a novel loss function which is termed as SoftmaxFocal loss is put forward to train the proposed network. The SoftmaxFocal loss creatively introduces an additive focusing term into original softmax loss which greatly enhances the online hard sample mining ability of the proposed model. Experiments conducted on three real datasets demonstrate the validity and superiority of the proposed model. [ABSTRACT FROM AUTHOR]

[Display omitted] Synthetic aperture radar (SAR) image change detection is a challenging task due to inherent speckle noise, imbalanced class occurrence and the requirement for discriminative feature learning. The traditional handcrafted feature extraction and current convolution-based deep learning techniques have some advantages, but suffer from being limited to neighborhood-based spatial information. The nonlocally observable imbalance phenomenon that exists naturally in small area change detection has presented a huge challenge to methods that focus on local features only. In this paper, an unsupervised method based on a variational graph auto-encoder (VGAE) network was developed for object-based small area change detection using SAR images, with the advantages of alleviating the negative impact of class imbalance and suppressing speckle noise. The main steps include: 1) Three types of difference image (DI) are combined to establish a three-channel fused DI (TCFDI), which lays the data-level foundation for subsequent analysis. 2) Simple linear iterative clustering (SLIC) is used to divide the TCFDI into superpixels regarded as nodes. Two functions are proposed and developed to measure the similarity between nodes to build a weighted undirected graph. 3) A VGAE network is designed and trained using the graph and nodes, and high-level nonlocal feature representations of each node are extracted. The network, with a Gaussian Radial Basis Function constrained by geospatial distances, establishes the connection among nonlocal, but similar superpixels in the process of feature learning, which leads to speckle noise suppression and distinguishable features learned in latent space. The nodes are then identified as changed or unchanged classes via k -means clustering. Five real SAR datasets were used in comparative experiments. Up to 99.72% accuracy was achieved, which is superior to state-of-the-art methods that pay attention only to local information, thus, demonstrating the effectiveness and robustness of the proposed approach. [ABSTRACT FROM AUTHOR]

In recent years, Convolutional Neural Networks (CNNs) have promoted the prosperity of Change Detection (CD). However, due to the intrinsic property of convolution kernel, this method cannot effectively model the long-distance dependency. The emergence of Vision Transformer (ViT) brings a new way to solve the problem. Based on ViT architecture, a novel Multi-Granularity remote sensing image Change Detection model (MGCDT) is proposed in this paper. We cascade several Local–Global Siamese Transformer (LGST) as backbone to extract local and global semantic discriminative features. In order to solve the serious problem of false detection and missing detection of feature boundary, a plug-and-play High Frequency Enhancement Unit (HFE) is proposed to replace the inflexible U-shaped structure to optimize the detection boundary. Considering the problem of multi-scale modeling of ground objects, a Multi-Scale Fusion Attention Unit (MSFA) is proposed, which integrates the flow of multi-scale information into the calculation process of self-attention. Finally, we utilize a Deep Feature Guidance Unit (DFG) to optimize the shallow detailed feature information. Extensive experiments show that, considering multi-granularity information, MGCDT outperforms the existing change detection algorithms on four remote sensing image change detection datasets. [ABSTRACT FROM AUTHOR]

With the rapid development of remote sensing data fusion technology, heterogeneous remote sensing image (HRSI) change detection (CD) has become a frontier field. The powerful nonlinear information processing capability of deep neural network provides the possibility for image domain conversion of HRSIs. However, the majority of existing methods rely on stacking network layers to extract deep high-level semantic features to accomplish the mutual conversion of image domains. The effective extraction and utilization of shallow features have not been fully considered. At the same time, the dependence between deep and shallow features has not been deeply explored. Therefore, these algorithms severely restrict the performance of CD. A new HRSI-CD network (DMF2Net) based on multi-level feature fusion is proposed in this paper to address these issues. DMF2Net uses central difference convolution to extract fine-grained features from shallow layers of images. It is capable of capturing the intrinsic detail features of image by aggregating intensity and gradient information. The dynamic multi-level feature fusion method is used to learn the fusion weights from the features, which are then used to guide the organic fusion of shallow and deep semantic features. This can preserve more position and detail features of the image, which can prevent the loss of small data during image conversion and enhance the model's ability to detect subtle changes. On this basis, a new DMF2Net-based method for detecting change in bi-temporal HRSIs is proposed. We conducted extensive experiments on four publicly available HRSI-CD datasets. The experimental results showed that the proposed CD framework achieves significant improvement over the most advanced methods. The project file of the proposed framework will be obtained from https://github.com/cwlnnu/DMF2Net. [ABSTRACT FROM AUTHOR]

Change detection using remote sensing images captured by small unmanned aerial vehicles (small UAVs) finds wide applications across various fields. However, there is a challenge when dealing with images captured at the same location by small UAVs at different times, leading to differences in viewpoint. These viewpoint differences present a significant challenge for most change detection methods. In this paper, we propose an end-to-end network, OFACD, designed to simultaneously address the issues of image alignment and change detection. Our network aligns feature maps using estimated optical flow and performs change detection concurrently. This approach enables the network to directly process images with viewpoint differences, effectively improving performance in scenarios with accumulated errors or large viewpoint variations, as well as enhancing throughput by eliminating repetitive feature extraction. Additionally, to fill the gap of the absence of change detection datasets with viewpoint differences and to evaluate our model, we created two change detection datasets with viewpoint differences. Extensive experimental results demonstrate that our method outperforms several state-of-the-art change detection methods in datasets involving viewpoint differences, exhibiting superior throughput and performance. • A novel change detection model for images with viewpoint differences is proposed. • OFACD simultaneously aligns images and detects changes toenhanceperformance. • Created two change detection datasets with viewpoint differences. • Extensive experimental results validate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The change detection task plays an irreplaceable role in the remote sensing field. However, most methods ignore the edge distinctive information. Such information is not only significant in some change detection tasks such as channel and river changes, but also important for refining the accuracy of change detection. Therefore, an edge refinement multi-feature (ERMF) extraction method, employing a siamese network to extract the primary discriminative features at five scales of bitemporal remote sensing images, is proposed in this paper. On the one hand, an edge refinement module is designed to obtain the edge change map as well as the final accurate region change map. On the other hand, a multi-level feature extraction module is introduced to acquire a coarse change map consisting of low-level location information and high-level semantic information at five different scales. Besides, it is worth emphasizing that we present a hybrid loss to evaluate the ERMF model. Experiments demonstrate that the ERMF model outperforms seven state-of-the-art methods in both qualitative and quantitative evaluations. • In this paper, a novel end-to-end remote sensing change detection model ERMF is proposed, which extracts edge change information and region change information. The results of regional change based on edge change information improved in the objective index, and boundary division is obvious in subjective evaluation. • Aiming at getting edge change information maps and refining the final results, an edge refinement module (ERM) is introduced. • In addition, a decent hybrid loss function is designed to constrain the results of edge change detection and region change detection. [ABSTRACT FROM AUTHOR]

Let X = (X t) t ≥ 0 be a process behaving as a general increasing Lévy process (subordinator) prior to hitting a given unknown level m 0 , then behaving as another different subordinator once this threshold is crossed. This paper addresses the detection of this unknown threshold m 0 ∈ [ 0 , + ∞ ] from an observed trajectory of the process. These kind of model and issue are encountered in many areas such as reliability and quality control in degradation problems. More precisely, we construct, from a sample path and for each ε > 0 , a so-called detection level L ε by considering a CUSUM inspired procedure. Under mild assumptions, this level is such that, while m 0 is infinite (i.e. when no changes occur), its expectation E ∞ (L ε) tends to + ∞ as ε tends to 0, and the expected overshoot E m 0 ( [ L ε − m 0 ] +) , while the threshold m 0 is finite, is negligible compared to E ∞ (L ε) as ε tends to 0. Numerical illustrations are provided when the Lévy processes are gamma processes with different shape parameters. [ABSTRACT FROM AUTHOR]

Co-seismic landslide mapping after earthquake event is essential for emergency rescue, geohazard prevention, and post-disaster reconstruction. Most co-seismic landslide mapping is primarily achieved via field surveys or visual interpretation of remote sensing images. However, such methods are highly labor-intensive and time-consuming, particularly over large areas. This paper proposed an automated co-seismic landslide mapping approach, which has three main advantages compared to state-of-the-art methods. First, it removes the dependence on the manual labeling for training samples through an unsupervised change detection process. Second, the approach takes the lead in introducing multi-scale extended morphological profiles to comprehensively describe the characteristics of various landslides induced by earthquake. Third, it is also the first attempt to employ ensemble strategy in landslide identification, which integrates the advantages of different machine learning-based classifiers, further improving the accuracy of recognition. Three experiments were carried out through multi-temporal Sentinel-2 and PlanetScope images acquired in China and Haiti. The results demonstrated the effectiveness and superiority of the proposed approach compared to other methods, providing an effective solution for complicated co-seismic landslide mapping task in the future. [ABSTRACT FROM AUTHOR]

Ecocultural restoration involves the reciprocal repair of ecosystems and revitalization of cultural practices to enhance their mutual resilience to natural and anthropogenic disturbances and climate change stressors. Resilient ecocultural systems are adapted to retain structure and function in the face of disturbances that remain within historical ranges of severity. To assist in ecocultural restoration and management, understanding how a system has historically responded to different types of disturbances is therefore invaluable in understanding how social-ecological resilience can be maintained in the face of future stressors and disturbances. However, records of disturbances and ecocultural responses can be limited for certain landscapes and human communities. In this methods paper, we demonstrate a mixed-method process for integrating oral history, field-based knowledge, archival information, and historical and contemporary aerial images to gain insight into the changes on the Klamath River in Northern California from the 1940s through 2020. We georegistered historical imagery, quantified changes between land cover classes, and contextualized these classifications with qualitative assessments of changes in larger surrounding areas. By synthesizing these data sources with field measurements, mining and other land survey maps, timber management plans, fire and flood histories, and interviews with members of the Karuk Tribe, we were able to reconstruct the land use and land cover change histories at five sites. We noted that recovery of canopy cover from fire and logging practices was faster than for flood, which was faster than recovery from mining, consistent with the relative severity of likely soil disturbance. By combining different sources of information with complementary strengths, we were able to provide managers with site-specific information on recovery from different types of disturbance. Though this approach was labor-intensive, with emerging tools for supervised classification of high-resolution imagery, mixed-method analytical historical ecology could be applied more broadly, supporting ecocultural restoration on a larger scale. [Display omitted] • Reconstructing landscape histories helps understand ecosystem response to disturbance. • Combining qualitative and quantitative analysis helps reconstruct land-use. • Synthesizing many historical data sources improves ecocultural restoration planning. • Manual classification could improve with machine learning 'human-in-the-loop' tools. [ABSTRACT FROM AUTHOR]

• A change detection method is proposed with heterogeneous remote sensing images. • It takes advantage of Cycle-Consistent Adversarial Networks for data mining. • Copula mixture is constructed to model the dependence structure of the images. • Experiments on real datasets validate the superiority of the proposed method. In this paper, we consider the problem of change detection (CD) with two heterogeneous remote sensing (RS) images. For this problem, an unsupervised change detection method has been proposed recently based on the image translation technique of Cycle-Consistent Adversarial Networks (CycleGANs), where one image is translated from its original modality to the modality of the other image so that the difference map can be obtained by performing arithmetical subtraction. However, the difference map derived from subtraction is susceptible to image translation errors, in which case the changed area and the unchanged area are less distinguishable. To overcome the above shortcoming, we propose a new unsupervised copula mixture and CycleGAN-based CD method (COMIC), which combines the advantages of copula mixtures on statistical modeling and the advantages of CycleGANs on data mining. In COMIC, the pre-event image is first translated from its original modality to the post-event image modality. After that, by constructing a copula mixture, the joint distribution of the features from the heterogeneous images can be learnt according to quantitive analysis of the dependence structure based on the translated image and the original pre-event image, which are of the same modality and contain totally the same objects. Then, we model the CD problem as a binary hypothesis testing problem and derive its test statistics based on the constructed copula mixture. Finally, the difference map can be obtained from the test statistics and the binary change map (BCM) is generated by K-means clustering. We perform experiments on real RS datasets, which demonstrate the superiority of COMIC over the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Change detection is the amount of changes that can guide to more concrete understandings into essential method concerning land cover, land usage and ecological variations. This paper deals with an intelligent methodology to optimize the solution out of the solution space that improves the efficiency of the change detection process. To support the theme, an integrated semi-supervised method is designed with a focus on image fusion, semi supervised clustering and binary swarm based optimization. A new approach of handling fusion using sparse coding is referred to expand the amount of information. Using the extracted information, change detection process is carried out by a constrained clustering technique to provide a solution reflecting the level of changes occurred in the region of investigation. To improve the accuracy by proposing a global optimum solution, still the result is refined through binary swarm based optimization process and hence the results are accelerated towards the increased level of accuracy followed by which the change map is reconstructed to show case changes prominently. To determine the accurateness of the proposed methodology quantitative and qualitative analysis has been done with different datasets. The proposed method has been evaluated with existing techniques such as k-means, AKM, FCM, ECKM and ASCC to show the efficiency and proved to be the preeminent change detection methodology compared to the state-of-art methods. [ABSTRACT FROM AUTHOR]

Multi-temporal high spatial resolution earth observation makes it possible to detect complex urban land surface changes, which is a significant and challenging task in remote sensing communities. Previous works mainly focus on binary change detection (BCD) based on modern technologies, e.g., deep fully convolutional network (FCN), whereas the deep network architecture for semantic change detection (SCD) is insufficiently explored in current literature. In this paper, we propose a deep multi-task encoder-transformer-decoder architecture (ChangeMask) designed by exploring two important inductive biases: sematic-change causal relationship and temporal symmetry. ChangeMask decouples the SCD into a temporal-wise semantic segmentation and a BCD, and then integrates these two tasks into a general encoder-transformer-decoder framework. In the encoder part, we design a semantic-aware encoder to model the semantic-change causal relationship. This encoder is only used to learn semantic representation and then learn change representation from semantic representation via a later transformer module. In this way, change representation can constrain semantic representation during training, which introduces a regularization to reduce the risk of overfitting. To learn a robust change representation from semantic representation, we propose a temporal-symmetric transformer (TST) to guarantee temporal symmetry for change representation and keep it discriminative. Based on the above semantic representation and change representation, we adopt simple multi-task decoders to output semantic change map. Benefiting from the differentiable building blocks, ChangeMask can be trained by a multi-task loss function, which significantly simplifies the whole pipeline of applying ChangeMask. The comprehensive experimental results on two large-scale SCD datasets confirm the effectiveness and superiority of ChangeMask in SCD. Besides, to demonstrate the potential value in real-world applications, e.g., automatic urban analysis and decision-making, we deploy the ChangeMask to map a large geographic area covering 30 km2 with 300 million pixels. Code will be made available. [ABSTRACT FROM AUTHOR]

The coastline of the Nile Delta is a dynamic zone in Egypt that is received considerable attention. This region contains many major cities and it also contains a number of economically important harbors such as Damietta, Port Said and Alexandria. This research focuses attention on applying Geographical Information System (GIS) tool to assess prospective changes along the coast of the Nile Delta, Egypt during the past thirty-four years. In this paper, changes to the coastline were monitored using twelve scenes of Landsat images from 1984 to 2018. After the extraction of shorelines from Landsat images, Digital Shoreline Analysis System (DSAS) was used to calculate shoreline displacement rate. The results showed that about 44.2% of the study area was under erosion and 45% was under accretion while it was stable at 10.8% between 1984 and 2018. The results illustrated that Rosetta and Damietta promontories are displayed higher rates of erosion than other coastal regions of the Nile Delta. National measures must be implemented to protect the entire coast in threatened areas. [ABSTRACT FROM AUTHOR]

Climate change is the biggest challenge faced by the world. It has already started affecting the weather patterns leading to disruption of normal life. Detecting change helps to monitor and plan the Earth's resources in an efficient manner. An important factor in climate change is the change in the green cover. Non-availability of standard datasets and limited labeled data points makes it difficult to attain high accuracies in change detection. In this paper, we have proposed a modified ensemble of extreme learning machines (mAEELM) for change detection in green cover to increase the accuracy of the change detection process. It uses Extreme Learning Machine (ELM) as the base classifier. Different ELMs are trained with different configuration so that they have different learning capabilities. These ELMs are combined to create an ensemble and it is then adapted based on the accuracy of the individual ELMs. The ensemble is then pruned to eliminate the ELMs which are not contributing towards the overall result of the ensemble, to make it more efficient. The proposed algorithm has been applied for detecting change on two areas of Gandhuan, Punjab and Chaparkaura Kham, Uttar Pradesh, India. The algorithm shows an average accuracy of 97.8% on both the datasets. [ABSTRACT FROM AUTHOR]

• Instead of supervised methods, unsupervised adversarial learning is proposed to construct spectral mapping network for HS change detection, which not only breaks the dilemma of limited training datasets, but also opens up a more flexible way for HS change detection. • A new spectral distance constraint is enforced on the spectral mapping network to enhance the adaptability of the network to change detection tasks. Specifically, it is enforced the reconstruction space spectral constraint loss and latent space adversarial loss to acquire a better spectral reconstruction effect and make the deep latent space obey multivariate Gaussian distribution. • Considering that the changed targets often exist in connected pixel areas rather than several isolated pixels, we pay attention to the spatial correlation to design the spatial attribute optimization, thereby significantly improving our UHCD framework performance. Unlike the existing change detection approaches based on the multispectral (MS) image and synthetic aperture radar (SAR) image datasets, a novel unsupervised hyperspectral change detection (UHCD) framework is proposed in this paper. The UHCD framework is designed for hyperspectral images with high dimensions and low availability. This framework consists of two modules: the spectral mapping with adversarial learning and the discriminant analysis with spatial attribute optimization. In comparison with other advanced change detection methods, the proposed framework possesses three distinctive properties: (1) The unsupervised spectral mapping is leveraged to exploit underlying spectral features without the requirement of pseudo-training datasets in the change detection task; (2) We introduce spectral constraint loss into reconstruction space and adversarial loss into latent space to enhance the quality of the features extracted by the spectral mapping network; (3) Spatial attribute optimization uses the spatial correlation to further improve the performance of the proposed UHCD method. The experimental results on two real datasets show that the proposed UHCD achieves competitive performance. [ABSTRACT FROM AUTHOR]

With strong penetrability and high resolution, synthetic aperture radar (SAR) images have been widely used in remote sensing image change detection. With the essence of heuristics, metaheuristics are suitable for the needs of most real-life optimization problems according to the expected solution quality and allowable calculation time. To improve the accuracy of change detection, this paper proposes a novel efficient metaheuristic change detection procedure. More specifically, we develop a superpixel-based multiobjective change detection method based on superpixel-wise feature representation and self-adaptive neighborhood-based binary difference evolution. During the clustering for feature analysis, a multiobjective optimization problem (MOP) is modeled which take the likelihood and Bhattacharyya distance between changed and unchanged classes as two objective functions. To solve the MOP, this paper firstly applies the superpixel binary representation technique into the encoding process of multiobjective evolutionary algorithm so that the dimension of pixel coding space can be substantially reduced. Then a self-adaptive neighborhood-based binary differential evolution strategy is proposed to explore the optimal change detection strategy, where a novel mutation operator integrating neighborhood information is designed for the improvement of change detection performance. Experimental results on three real SAR datasets confirm that the proposed method can make a better performance on change detection and the metaheuristic algorithm has good convergence and stability. • Change detection of synthetic aperture radar images. • Superpixel-based difference feature representation strategy. • Difference analysis in a form of evolutionary multi-objective optimization. • A self-adaptive neighborhood-based binary difference evolution framework. [ABSTRACT FROM AUTHOR]

This paper proposes a change detection algorithm based on Fully Convolutional Siamese Networks for optical aerial images, which is trained by using an improved Contrastive Loss-Focal Contrastive Loss (FCL). The proposed framework equipped with contrastive loss can extract features directly from image pairs and measure changes by using a distance metric. In other words, this method encourages reducing intra-class variance and enlarging inter-class difference, so that the binarized change map can be obtained by a simple threshold. In change detection task, a critical problem is how to overcome example imbalance (i.e. unchanged examples are much more than changed examples). To address this challenge, a novel focal contrastive loss is proposed to further improve the performance of the model. FCL can reduce the impact of example imbalance and make the model focus learning on hard examples. Extensive experiments demonstrate that the proposed approach is more abstract as well as robust. Compared with other baseline methods, the presented method achieves better results on SAZDA, TISZADOB, CDD, and WHU-CD data set. It achieves state-of-the-art performance in terms of the F1 measure. [ABSTRACT FROM AUTHOR]

• Analyzed areas of tree removal along interstates in Georgia, USA with NAIP imagery. • 25% of previously forested interstate roadsides to have experienced tree removal. • Tree crashes were not reduced despite this program goal. • Agencies need to consider measures other than tree removal to prevent tree crashes. 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. [ABSTRACT FROM AUTHOR]

Cyber-resilience is an increasing concern for autonomous navigation of marine vessels. This paper scrutinizes cyber-resilience properties of marine navigation through a prism with three edges: multiple sensor information fusion, diagnosis of not-normal behaviours, and change detection. It proposes a two-stage estimator for diagnosis and mitigation of sensor signals used for coastal navigation. Developing a Likelihood Field approach, the first stage extracts shoreline features from radar and matches them to the electronic navigation chart. The second stage associates buoy and beacon features from the radar with chart information. Using real data logged at sea tests combined with simulated spoofing, the paper verifies the ability to timely diagnose and isolate an attempt to compromise position measurements. A new approach is suggested for high level processing of received data to evaluate their consistency, which is agnostic to the underlying technology of the individual sensory input. A combined generalized likelihood ratio test using both parametric Gaussian modelling and Kernel Density Estimation is suggested and compared with a detector using only either of two. The paper shows how the detection of deviations from nominal behaviour is possible when the navigation sensor is under attack or defects occur. • Erroneous information from navigation sensors is identified and mitigated as follows. • A two-stage estimator is designed for coastal navigation to validate Radar and Gyro. • Radar and ENC shoreline features are first matched in a Likelihood Field approach. • Thereafter, buoy and beacon features are matched to isolate defects in information. • Parametric and non-parametric models are employed for efficient GLRT change detection. • Simulating spoofing attacks as overlay on sea test data demonstrates resilience. [ABSTRACT FROM AUTHOR]

Monitoring forest cover change from Earth observation data streams in near-real-time presents a challenge for automated change detection by way of a continuously updated big dataset. Even though deforestation is a significant global problem, forest cover changes in pairs of subsequent images happen relatively infrequently. Detecting a change can require the download and processing of tens, hundreds or even thousands of images. In geoscientific applications of Earth observation, machine learning algorithms are increasingly used. Once trained, a machine learning model can be applied to new images automatically. This paper introduces the open-access Python 3 package Pyeo - " Py thon for E arth O bservation". Pyeo provides a set of portable, extensible and modular Python functions for the automation of machine learning applications from Earth observation data streams, including automated search and download functionality, pre-processing and atmospheric correction, re-projection, creation of thematic base layers and machine learning classification or regression. Pyeo enables users to train their own machine learning models and then apply the models to newly downloaded imagery over their area of interest. This paper describes in detail how Pyeo works, its requirements, benefits, and a description of the libraries used. An application to the automated forest cover change detection in a region in Kenya is given. Pyeo can be used on cloud computing architectures such as Amazon Web Services, Microsoft Azure and Google Colab to provide scalable applications and processing solutions for the geosciences. • Highlight 1: A Python package for Earth observation processing chains for change detection is presented. • Highlight 2: Data can be processed in near-real-time whenever a new satellite image is acquired. • Highlight 3: The satellite change detection algorithm informs the user of detected change events. • Highlight 4: An application to a forest in Kenya is presented to demonstrate the software. • Highlight 5: This software is used by the Kenya Forest Service for monitoring deforestation from Sentinel-2. [ABSTRACT FROM AUTHOR]

• Proposes an end-to-end network for biphasic image change detection. • Can better solve the problem of detection errors. • High degree of automation in the calculation process. • Enhanced feature extraction module was inroduced. Change detection is an effective means of monitoring surface changes. In order to solve the problem of reducing detection accuracy in remote sensing image change detection methods that only focus on extracting deep semantic information while ignoring high-resolution shallow information, this paper proposes an end-to-end multi-side output fusion deep supervised network for high-resolution biphasic remote sensing image change detection. The enhanced feature extraction module is used to increase the Receptive field and capture multi-scale features. The dense skip connection module utilizes skip connections to fuse shallow position information and deep semantic information, alleviating the problem of feature information loss. The deep supervision module with multiple side-outputs fusion utilizes the information complementarity between different scale feature maps, combined with multi-scale prediction maps, to improve the robustness of target scale changes. The combined Loss function is used to alleviate the problem of imbalance between positive and negative samples, so that more attention is paid to the learning of changing characteristics in the process of training the network. The proposed method was validated on CDD dataset and DSIFN dataset, with F1 scores of 91.58% and 86.69%, respectively, which were improved by 1.00% and 5.53% compared to the suboptimal network. [ABSTRACT FROM AUTHOR]

Detecting fine-grained changes among a set of images taken at different times is a challenging problem, which is important for the applications such as high-value scene monitoring and structural inspection. Existing methods use a geometry registration and lighting compensation pipeline to alleviate the ingredients that affect change detection. However, lighting compensation may introduce a lot of artifacts, e.g., inaccurate local lighting conditions and fine-grained change disappearance. The distortion introduced by general registration methods also makes the process of lighting correction more challenging. To solve these problems, in this paper, we propose a new phase-based change detection method, called as pb-FGCD, which first aligns two observation of different times and then extract structure information to handle lighting difference. Benefiting from the proposed weighted amplitude filter and iterative alignment strategy, we could obtain a better registration performance. The structure information extracts the lighting irrelevant components of an image and is more sensitive to structure changes compared with the lighting compensation based method, which is vital for detecting the changes with low intensity difference. Furthermore, we contribute a new change detection dataset, which contains 100 groups of the real-world data of the ancient Mogao murals and the Terra-Cotta. It is the first fine-grained change detection dataset focusing on the ancient mural. Our approach significantly improves the accuracy by more than 190% higher F1-measure than the state-of-the-art methods. • A novel fine-grained change detection framework. • A novel phase-based fine-grained registration method. • Finding structural information effective for fine-grained change detection. • Building a real-world dataset for fine-grained change detection. • Increasing the F1-measure by more than 190% over those state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Change detection is important in remote sensing image analysis. In recent years, significant breakthroughs have been made in change detection algorithms based on deep learning. However, due to continuous downsampling, the detection results of these algorithms still have serious detection errors, detection omissions and edge blurring. Aiming at these problems, this paper proposes a dual-branch network for change detection. The network has two branches, which are used to extract the depth-variant semantic features of the multi-temporal image pairs and the respective features of each image respectively. In addition, we designed a Multi-scale Strip Convolution Module (MSCM) to extract the multi-scale features of the image, a new Spatial Attention Module (SAM) to strengthen the feature representation of changing regions, and a Feature Fusion Network (FFN) to guide the fusion between multiple features of the two branches. Experimental results show that the proposed method substantially mitigates detection errors, detection omissions and obtains sharper edges, it outperforms other current algorithms. [ABSTRACT FROM AUTHOR]

This paper presents a new system to detect changes between reference and target videos suitable for small-scale datasets. Twin pre-trained ResNet-50 features are processed using a learning-based pipeline that has a limited number of adjustable parameters, allowing end-to-end training even on relatively small databases. This is achieved with two innovative modules in tandem: a low-complexity dissimilarity module and a post-processing step using learnable morphological operations. Both can be smoothly incorporated in optimization procedures that employ gradient-based algorithms. The pipeline ends with temporal consistency and change classification modules, and it is evaluated on the VDAO dataset, a challenging database of videos recorded with moving cameras in a cluttered industrial environment. Ablation studies show how each proposed module contributes to the final system performance, with a prominence role for the newly proposed ones. Results indicate that the proposed system achieves a detection performance that is about 18% superior to the one of current state-of-the-art methods. Software, results, and a pre-trained architecture of the proposed framework are available at https://github.com/rafaelpadilla/TCF-LMO. [Display omitted] • Change-detection system using deep-learning features extracted from twin networks. • Proposed model developed to allow training on small datasets. • Novel dissimilarity module provides a flexible use of reference and target features. • Innovative morphological operations trained in a gradient-based learning framework. • Detection performance 18% superior to the one of current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• We present a new paradigm that reduces change detection to semantic segmentation which means tailoring an existing and powerful semantic segmentation network to solve change detection. • Our analysis suggests that there are three possible change types within the change detection task and they should be learned separately. Therefore, we propose a MTF (Merge Temporal Features) module to learn these changes. • We propose a simple but effective network, called C-3PO (Combine 3 POssible change types), detects changes in pixel-level, and can be considered as a new baseline network in this field. Change detection (CD) aims to identify changes that occur in an image pair taken different times. Prior methods devise specific networks from scratch to predict change masks in pixel-level, and struggle with general segmentation problems. In this paper, we propose a new paradigm that reduces CD to semantic segmentation which means tailoring an existing and powerful semantic segmentation network to solve CD. This new paradigm conveniently enjoys the mainstream semantic segmentation techniques to deal with general segmentation problems in CD. Hence we can concentrate on studying how to detect changes. We propose a novel and importance insight that different change types exist in CD and they should be learned separately. Based on it, we devise a module named MTF to extract the change information and fuse temporal features. MTF enjoys high interpretability and reveals the essential characteristic of CD. And most segmentation networks can be adapted to solve the CD problems with our MTF module. Finally, we propose C-3PO, a network to detect changes at pixel-level. C-3PO achieves state-of-the-art performance without bells and whistles. It is simple but effective and can be considered as a new baseline in this field. Our code for C-3PO is available at https://github.com/DoctorKey/C-3PO. [ABSTRACT FROM AUTHOR]

• Detection of cropland abandonment based on phenological information. • Summarized the inter-annual phenological changes of abandoned cropland. • Detecting cropland abandonment information at the sub-pixel scale. • Spatiotemporal patterns of abandoned cropland in this paper were revealed. Cropland abandonment is a common land-use change with mixed impacts on the environment and rural economic development. Prevalent small family farms and excessive land fragmentation result in cropland abandonment processes that are often gradual and spatially dispersed. These difficulties limit the ability to apply previous remote sensing mapping methods of cropland abandonment to areas with complex underlying surfaces, where mixed pixels and spectral aliasing problems arise. To meet the demands of broad-scale and high-accuracy abandoned cropland mapping, we developed the P C R R S B S - C V model by combining the phenology-based cropland retirement remote sensing (PCRRS) model with the coefficient of variation (CV) and tilled soil fraction (BS). This method combines information regarding crop phenology with Moderate Resolution Imaging Spectroradiometer (MODIS) time-series images. The PCRRS model was used to estimate changes in spectral metrics parameters during the process of cropland abandonment. Additionally, the dead fuel index and normalized difference vegetation index (NDVI) were employed to estimate the BS in mixed pixels (according to crop type). We predicted that use of the BS as the weighting coefficient for the PCRRS model would reduce interference from the mixed pixel problem, while spatial heterogeneity could be reduced by dividing the research area into regional units. Use of the CV of the NDVI time series with phenological information highlights the volatility of crop growth periods and helps to eliminate disturbances associated with woodlands and grasslands. Finally, we demonstrated this method in the Loess Plateau of China and a portion of central Europe; we verified its accuracy using high-resolution images from Google Earth. Our algorithm demonstrated overall accuracy of 82.2 % and can extract cropland abandonment information as little as 20 % of a pixel. The overall root mean square error (RMSE) was controlled below 15 %. In summary, the high accuracy achieved by this method enables the monitoring of cropland abandonment dynamics on a large scale. [ABSTRACT FROM AUTHOR]

Remote sensing image change detection remains a challenging task. Most existing approaches are based on fully supervised learning, but labeled data are so scarce for change detection. It is difficult to exhibit high detection performance with a limited amount of labeled data. In this paper, we propose a semi-supervised Label Propagation (SSLP) approach for multi-source remote sensing image change detection. First, a clustering label propagation (CLP) method is designed to cluster pre and post images, respectively, and assign pseudo labels to unlabeled pixel pairs that have similar mapping relationships to labeled pixel pairs. Second, a pixel density metric is investigated to filter out the data with low density and retain the data with high density, which can ensure the reliability of the propagated data. Third, a secondary expansion method based on pixel neighborhood is used to generate enough training data for training a classifier. Finally, the effectiveness of SSLP is validated on three real datasets by comparing to other related methods. • Propose a clustering label propagation method to expand the labeled data. • Propose a density-based denoising method to ensure the reliability of the propagated data. • Apply the proposed approach to change detection from multi-source remote sensing images. • Numerous experiments demonstrate the effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

• A novel supervised domain adaptation framework SDACD for cross-domain change detection. • SDACD unifies image adaptation and feature adaptation in an end-to-end trainable manner. • SDACD can handle cross-domain change detection and consistently improve the performance as an easy-to-plug-in module. • Our SNUNet-based framework sets new state-of-the-art performance with an F1-score of 97.34% on CDD dataset and 92.36% on WHU building dataset. Change detection is a crucial but extremely challenging task in remote sensing image analysis, and much progress has been made with the rapid development of deep learning. However, most existing deep learning-based change detection methods try to elaborately design complicated neural networks with powerful feature representations. However, they ignore the universal domain shift induced by time-varying land cover changes, including luminance fluctuations and seasonal changes between pre-event and post-event images, thereby producing suboptimal results. In this paper, we propose an end-to-end supervised domain adaptation framework for cross-domain change detection named SDACD, to effectively alleviate the domain shift between bi-temporal images for better change predictions. Specifically, our SDACD presents collaborative adaptations from both image and feature perspectives with supervised learning. Image adaptation exploits generative adversarial learning with cycle-consistency constraints to perform cross-domain style transformation, which effectively narrows the domain gap in a two-side generation fashion. As for feature adaptation, we extract domain-invariant features to align different feature distributions in the feature space, which could further reduce the domain gap of cross-domain images. To further improve the performance, we combine three types of bi-temporal images for the final change prediction, including the initial input bi-temporal images and two generated bi-temporal images from the pre-event and post-event domains. Extensive experiments and analyses conducted on two benchmarks demonstrate the effectiveness and generalizability of our proposed framework. Notably, our framework pushes several representative baseline models up to new State-Of-The-Art records, achieving 97.34% and 92.36% on the CDD and WHU building datasets, respectively. The source code and models are publicly available at https://github.com/Perfect-You/SDACD. [ABSTRACT FROM AUTHOR]

Remote sensing image change detection (CD) technology is an important means to understand the changes of the earth surface. In recent years, many excellent CD methods based on convolutional neural networks (CNN) have been proposed. Although these methods have achieved good results, CD is still a very challenging task. Recently, many visual algorithms based on multi-layer perceptron (MLP) have been proposed and achieved similar or even better results than CNN-based algorithms in many visual fields. Inspired by the success of these MLP-based algorithms, in this paper we propose a siamese U-shaped MLP-based CD network named SUMLP. First, SUMLP uses CycleMLP block as the basic unit of the network and combines the characteristics of the CD task to construct a MLP-based CD network with siamese U-shaped structure. Secondly, to better optimize the CD network, we design a new loss function, weighted cosine cross entropy (WCCE) loss. WCCE loss has two advantages: 1. WCCE loss gives different weights to the changed samples and the unchanged samples, which can cope with the data imbalance problem in the CD task; 2. WCCE loss uses cosine margin to increase the inter class distance and the intra class compactness, so as to better distinguish the changed samples from the unchanged samples. For comprehensively verifying the validity of our proposed method, we carried out many experiments on three public CD datasets: LEVIR-CD dataset, CDD dataset and WHU-CD dataset. On these datasets, SUMLP's F 1 − s c o r e are at least 0.33%, 1.41% and 3.56% higher than the other relevant CD methods respectively. • A Siamese U-shaped MLP-based Network, SUMLP, was proposed, which is the first CD network based entirely on MLP. • In order to better optimize the CD network, weighted cosine cross entropy (WCCE) loss was designed. • The effectiveness of SUMLP and WCCE loss was evaluated. [ABSTRACT FROM AUTHOR]

The study of elevation change using pre- and post-event raster grids obtained by airborne laser scans is central to many scientific areas, including geomorphology, landscape science and environmental science. Separating real elevation change from noise is a nontrivial task due to complex sources of measurement uncertainty causing errors to be spatially variable. For high-resolution grids, an additional hurdle is the excessive computational effort required by common statistical models for the spatial error. This paper introduces a screening methodology for change detection in high-resolution elevation difference grids using a composite likelihood regularization framework. We propose to model the spatial error by a sparse spatial autoregressive conditional heteroscedasticity process and obtain simultaneous detection of change location and change size estimation by maximizing a penalized composite likelihood objective. The methodology is implemented through a fast coordinate-wise maximization algorithm with computational complexity growing linearly with the grid size. Numerical studies on simulated and real data show satisfactory accuracy for both change detection and parameter estimation. [ABSTRACT FROM AUTHOR]

Automatic construction progress documentation and metric evaluation of execution work in confined building interiors requires particularly reliable geometric evaluation and interpretation of statistically uncertain as-built point clouds. This paper presents a method for high-resolution change detection based on dense 3D point clouds from terrestrial laser scanning (TLS) and the discretization of space by voxels. In order to evaluate the metric accuracy of a BIM according to the Level of Accuracy (LOA) specification, the effects of laser range measurements on the occupancy of space are modeled with belief functions and evaluated using Dempster and Shafer's theory of evidence. The application is demonstrated on the point cloud data of multi temporal scanning campaigns of real indoor reconstructions. The results show that TLS point clouds are suitable to verify a given BIM up to LOA 40 if special attention is paid to the scanning geometry during the acquisition. The proposed method can be used to document construction progress, verify and even update the LOA status of a given BIM, confirming valid and BIM-compliant as-built models for further planning. • A new method for fusion of model uncertainty with uncertainties of point clouds. • Consideration of indoor scanning geometry for accuracy assessment. • Voxel based change detection and self-occlusion analysis. • Experiments on progress evaluation on two real construction sites. [ABSTRACT FROM AUTHOR]

• Estimates obtained via multiplicative updates are consistent with diffusion. • Stationarity of estimates is a promising indicator for learning progress. • Utilization of a dynamic learning rate in learning under concept drift is beneficial. Dynamic systems are highly complex and hard to deal with due to their subject- and time-varying nature. The fact that most of the real world systems/events are of dynamic character makes modeling and analysis of such systems inevitable and charmingly useful. One promising estimation method that is capable of unlearning past information to deal with non-stationarity is Stochastic Learning Weak Estimator (SLWE) by Oommen and Rueda (2006). However, due to using a constant learning rate, it faces a trade-off between plasticity and stability. In this paper, we model SLWE as a random walk and provide rigorous theoretical analysis of asymptotic behavior of estimates to obtain a statistical model. Utilizing this model, we detect changes in stationarity to switch between exploratory and exploitative learning modes. Experimental evaluations on both synthetic and real world data show that the proposed method outperforms related algorithms in different types of drifts. [ABSTRACT FROM AUTHOR]

Structure from motion (SfM) photogrammetry is an increasingly common technique for measuring landscape change over time by deriving 3D point clouds and surface models from overlapping photographs. Traditional change detection approaches require photos that are geotagged with a differential GPS (DGPS) location, which requires expensive equipment that can limit the ability of communities and researchers to perform frequent (i.e. daily, weekly, and/or monthly) surveys. Crowd-sourced photos can lower the barrier to entry and substantially increase the frequency of surveys, although such photos often lack accurate location information and can vary in quality. This paper presents a SfM approach for monitoring environmental change in high relief coastal environments that does not require all photos have DGPS location information and does not require field survey data. A 1.5 km section of coastal bluffs near the Elwha River Delta (Washington state) is used to demonstrate the efficacy of this approach. Photos of the bluff were collected with a digital SLR camera or phone camera while either on foot along the beach or from a boat as part of monitoring following removal of two large dams along the Elwha River during 2011–2013. Only 33% of photos had DGPS location information, whereas most photos had no location information or locations that were accurate to a couple of meters. All photos were processed using 3D, 4D, and fixed-floating (FF) SfM alignment methods and the resulting dense point clouds are used to compare the different alignment approaches with crowd-sourced photo sets. Results demonstrate that 4D and FF approaches are more likely to reconstruct and are more accurate than the 3D approach. While the 4D and FF have comparable accuracies, the FF approach is several orders of magnitude more efficient, as this method can leverage camera location information from relatively few photos to improve the accuracy of all aligned and derived products. Effectively utilizing crowd-sourced photos in SfM change detection can improve the frequency of surveying a landscape in a more cost-effective approach that also has potential for citizen-science engagement and communication. This is especially important for data-poor environments such as high-relief coastal cliffs and bluffs, where near-nadir imagery and LIDAR may fail to accurately capture near-vertical cliffs or bluff faces. Based on the analysis of different photo alignment and filtering approaches, we present suggested best practices for engaging citizen scientists in coastal cliff and bluff monitoring efforts through collecting photos amenable for SfM reconstruction. • SfM and crowd-sourced photos can fill data gaps in coastal research. • Differential GPS information is not required for accurate point cloud data. • Minimal training is required for photographers to contribute to research. • A list of best practices for collecting photos can help guide training. • Engaging stakeholders to collect photos is more efficient than regular surveys. [ABSTRACT FROM AUTHOR]

The high spectral resolution of hyperspectral image (HSI) provides the possibility to capture the subtle changes associated with land-cover dynamic evolution process. Supervised deep leaning approaches have been extensively applied to HSI change detection task. However, their success is attributed to the large amount of annotated training samples. Moreover, the large receptive field in the convolutional layer and the presence of the pooling layer reduces the spatial resolution of the deepest FCN layer which will make the predicted change map tends to lack fine object boundary details. In order to boost the HSI change detection performance in a reduced labor of annotating data and enhance the object boundary details of the change map, in this paper, we propose a novel HSI change detection method that integrates both iterative active learning and affinity graph learning into a unified framework. The proposed method consists of two major branches, a unary HSI change detection network branch which learns the pixel-wise change probability, and a pairwise HSI affinity graph learning branch that learns the pairwise affinity of the hyperspectral difference image to refine the coarse probabilities through a random walk connection. To actively select the most informative unlabeled samples, we also propose an HSI change detection active learning strategy based on the spectral property of HSI difference image and the refined change probabilities. The procedures are conducted iteratively to obtain better change detection results progressively. Experimental results show that the proposed method can extract accurate subtle change information while properly preserving the edges and textures of the HSIs with significantly fewer labeled data. • A novel two branch active learning framework is proposed for hyperspectral image change detection. • The unary HSI change detection network branch learns the coarse change probabilities. • The pairwise HSI affinity graph learning branch is utilized to refine the coarse probabilities. • HSI change detection active learning strategy is proposed to actively select the most informative samples. • The procedures are conducted iteratively to obtain better change detection results. [ABSTRACT FROM AUTHOR]

Time series of networks are increasingly prevalent in modern data and pose unique challenges to pattern extraction and change detection. In this paper we develop and present a novel methodology to detect regime changes within a sequence of networks that have overlapping and evolving community structure. The core of the methodology is a non-negative matrix factorization that maximizes a Poisson likelihood subject to a penalty that accounts for sparsity in the network. By fitting the factorization model over a rolling window with a fast numerical optimization algorithm, change detection is accomplished by statistical monitoring of the matrix factors' evolution. A novel statistic is used to characterize the overall network evolution as well as the contribution of each node to the change. We demonstrate that the proposed methodology compares favorably with alternative techniques for on-the-go network change detection using synthetic and real data. A detailed case study on the 2007–2009 financial crisis and the European sovereign debt crisis shows the promise of the methodology for regulators as it identifies particular banks that contributed to each crisis in addition to identifying changing market conditions. • We develop a new methodology for detecting changes in core–periphery networks. • Matrix factorization and a novel change statistic characterize network evolution. • EWMA control charts are used for monitoring and change detection. • We validate the method using simulated and real interbank lending data. [ABSTRACT FROM AUTHOR]

As the key to the construction progress monitoring, methods and strategies for change detection using 3D point clouds from various sources have been investigated for years. However, how to achieve object-level change detection with uncertainty evaluation is still an unsolved topic. Occlusions and noise in 3D points and other attribute information, such as colors, lead to problems in the task of change detection. In this paper, we present a semantic-aided change detection method aimed at monitoring construction progress using UAV-based photogrammetric point clouds. Our framework consists of two key parts, which identify changes in a progressive manner: The first part consists of the detection of geometric changes using occupancy-based spatial difference identification, which indicates the changes of occupancy in 3D space, comprising changes in appearance or shape of building objects. In occupancy-based change detection, occupancy conflicts of occupied space and empty space along the viewing rays of cameras can be detected by considering the sensor positions. At the same time, occlusions can be handled implicitly. The second part involves changes in semantics, which are used to detect changes where the occupancy-based change detection is not sufficient for presenting changes, due to limitations of parameter settings or lack of attribute information. By utilizing semantic segmentation results presented by class probabilities, the uncertainty of the semantic changes can be estimated. For the detection of geometric and semantic changes, Dempster–Shafer theory is applied to fuse information from data acquired in different time epochs to detect changes. Using the two different types of changes, we can fully consider the changes that may happen at the construction sites and define the differences between the changes. By utilizing the proposed change detection methods, changes with different characteristics, including geometric changes and semantic changes, can be correctly identified. In a specific example, for a construction period from Dec 12, 2014 to Jan 16, 2015, 97.8% of the changed areas could be successfully detected. For the other construction period, from Jan 16, 2015 to Feb 26, 2015, 93.6% of changes were correctly detected. • A change detection framework using UAV image-based photogrammetric point clouds is proposed. • Observation limitations are addressed by exploiting occupancy conflicts that retrieve viewing rays of images. • A semantic-aided change detection method is proposed to achieve object level change analysis. • A case study in a real construction site using UAV-based photogrammetric point clouds is presented. [ABSTRACT FROM AUTHOR]

• Deep object-based semantic change detection framework (ChangeOS) is proposed. • ChangeOS seamlessly integrates object-based image analysis and deep learning. • City-scale building damage assessment can be achieved within one minute. • A global-scale dataset is used to evaluate the effectiveness of ChangeOS. • Two local-scale datasets are used to show its great generalization ability. Sudden-onset natural and man-made disasters represent a threat to the safety of human life and property. Rapid and accurate building damage assessment using bitemporal high spatial resolution (HSR) remote sensing images can quickly and safely provide us with spatial distribution information and statistics of the damage degree to assist with humanitarian assistance and disaster response. For building damage assessment, strong feature representation and semantic consistency are the keys to obtaining a high accuracy. However, the conventional object-based image analysis (OBIA) framework using a patch-based convolutional neural network (CNN) can guarantee semantic consistency, but with weak feature representation, while the Siamese fully convolutional network approach has strong feature representation capabilities but is semantically inconsistent. In this paper, we propose a deep object-based semantic change detection framework, called ChangeOS, for building damage assessment. To seamlessly integrate OBIA and deep learning, we adopt a deep object localization network to generate accurate building objects, in place of the superpixel segmentation commonly used in the conventional OBIA framework. Furthermore, the deep object localization network and deep damage classification network are integrated into a unified semantic change detection network for end-to-end building damage assessment. This also provides deep object features that can supply an object prior to the deep damage classification network for more consistent semantic feature representation. Object-based post-processing is adopted to further guarantee the semantic consistency of each object. The experimental results obtained on a global scale dataset including 19 natural disaster events and two local scale datasets including the Beirut port explosion event and the Bata military barracks explosion event show that ChangeOS is superior to the currently published methods in speed and accuracy, and has a superior generalization ability for man-made disasters. [ABSTRACT FROM AUTHOR]