Showing papers on "Change detection published in 2021"
TL;DR: The weighted double-margin contrastive loss is proposed to address the imbalanced sample is a serious problem in change detection, i.e., unchanged samples are much more abundant than changed samples, which is one of the main reasons for pseudochanges.
Abstract: Change detection is a basic task of remote sensing image processing. The research objective is to identify the change information of interest and filter out the irrelevant change information as interference factors. Recently, the rise in deep learning has provided new tools for change detection, which have yielded impressive results. However, the available methods focus mainly on the difference information between multitemporal remote sensing images and lack robustness to pseudochange information. To overcome the lack of resistance in current methods to pseudochanges, in this article, we propose a new method, namely, dual attentive fully convolutional Siamese networks, for change detection in high-resolution images. Through the dual attention mechanism, long-range dependencies are captured to obtain more discriminant feature representations to enhance the recognition performance of the model. Moreover, the imbalanced sample is a serious problem in change detection, i.e., unchanged samples are much more abundant than changed samples, which is one of the main reasons for pseudochanges. We propose the weighted double-margin contrastive loss to address this problem by punishing attention to unchanged feature pairs and increasing attention to changed feature pairs. The experimental results of our method on the change detection dataset and the building change detection dataset demonstrate that compared with other baseline methods, the proposed method realizes maximum improvements of 2.9% and 4.2%, respectively, in the F 1 score. Our PyTorch implementation is available at https://github.com/lehaifeng/DASNet .
324 citations
TL;DR: HuHuang et al. as discussed by the authors proposed a bitemporal image transformer (BIT) to efficiently and effectively model contexts within the spatial-temporal domain, and incorporated BIT in a deep feature differencing-based CD framework.
Abstract: Modern change detection (CD) has achieved remarkable success by the powerful discriminative ability of deep convolutions. However, high-resolution remote sensing CD remains challenging due to the complexity of objects in the scene. Objects with the same semantic concept may show distinct spectral characteristics at different times and spatial locations. Most recent CD pipelines using pure convolutions are still struggling to relate long-range concepts in space-time. Nonlocal self-attention approaches show promising performance via modeling dense relationships among pixels, yet are computationally inefficient. Here, we propose a bitemporal image transformer (BIT) to efficiently and effectively model contexts within the spatial-temporal domain. Our intuition is that the high-level concepts of the change of interest can be represented by a few visual words, that is, semantic tokens. To achieve this, we express the bitemporal image into a few tokens and use a transformer encoder to model contexts in the compact token-based space-time. The learned context-rich tokens are then fed back to the pixel-space for refining the original features via a transformer decoder. We incorporate BIT in a deep feature differencing-based CD framework. Extensive experiments on three CD datasets demonstrate the effectiveness and efficiency of the proposed method. Notably, our BIT-based model significantly outperforms the purely convolutional baseline using only three times lower computational costs and model parameters. Based on a naive backbone (ResNet18) without sophisticated structures (e.g., feature pyramid network (FPN) and UNet), our model surpasses several state-of-the-art CD methods, including better than four recent attention-based methods in terms of efficiency and accuracy. Our code is available at https://github.com/justchenhao/BIT_CD.
290 citations
TL;DR: Experimental results show that the proposed SNUNet-CD method improves greatly on many evaluation criteria and has a better tradeoff between accuracy and calculation amount than other state-of-the-art (SOTA) change detection methods.
Abstract: Change detection is an important task in remote sensing (RS) image analysis. It is widely used in natural disaster monitoring and assessment, land resource planning, and other fields. As a pixel-to-pixel prediction task, change detection is sensitive about the utilization of the original position information. Recent change detection methods always focus on the extraction of deep change semantic feature, but ignore the importance of shallow-layer information containing high-resolution and fine-grained features, this often leads to the uncertainty of the pixels at the edge of the changed target and the determination miss of small targets. In this letter, we propose a densely connected siamese network for change detection, namely SNUNet-CD (the combination of Siamese network and NestedUNet). SNUNet-CD alleviates the loss of localization information in the deep layers of neural network through compact information transmission between encoder and decoder, and between decoder and decoder. In addition, Ensemble Channel Attention Module (ECAM) is proposed for deep supervision. Through ECAM, the most representative features of different semantic levels can be refined and used for the final classification. Experimental results show that our method improves greatly on many evaluation criteria and has a better tradeoff between accuracy and calculation amount than other state-of-the-art (SOTA) change detection methods.
256 citations
TL;DR: A deeply supervised (DS) attention metric-based network (DSAMNet) is proposed in this article to learn change maps by means of deep metric learning, in which convolutional block attention modules (CBAM) are integrated to provide more discriminative features.
Abstract: Change detection (CD) aims to identify surface changes from bitemporal images. In recent years, deep learning (DL)-based methods have made substantial breakthroughs in the field of CD. However, CD results can be easily affected by external factors, including illumination, noise, and scale, which leads to pseudo-changes and noise in the detection map. To deal with these problems and achieve more accurate results, a deeply supervised (DS) attention metric-based network (DSAMNet) is proposed in this article. A metric module is employed in DSAMNet to learn change maps by means of deep metric learning, in which convolutional block attention modules (CBAM) are integrated to provide more discriminative features. As an auxiliary, a DS module is introduced to enhance the feature extractor's learning ability and generate more useful features. Moreover, another challenge encountered by data-driven DL algorithms is posed by the limitations in change detection datasets (CDDs). Therefore, we create a CD dataset, Sun Yat-Sen University (SYSU)-CD, for bitemporal image CD, which contains a total of 20,000 aerial image pairs of size 256 x 256. Experiments are conducted on both the CDD and the SYSU-CD dataset. Compared to other state-of-the-art methods, our network achieves the highest accuracy on both datasets, with an F1 of 93.69% on the CDD dataset and 78.18% on the SYSU-CD dataset.
206 citations
TL;DR: Wang et al. as discussed by the authors proposed a bitemporal image transformer (BIT) to efficiently and effectively model contexts within the spatial-temporal domain, where the high-level concepts of the change of interest can be represented by a few visual words.
Abstract: Modern change detection (CD) has achieved remarkable success by the powerful discriminative ability of deep convolutions. However, high-resolution remote sensing CD remains challenging due to the complexity of objects in the scene. Objects with the same semantic concept may show distinct spectral characteristics at different times and spatial locations. Most recent CD pipelines using pure convolutions are still struggling to relate long-range concepts in space-time. Non-local self-attention approaches show promising performance via modeling dense relations among pixels, yet are computationally inefficient. Here, we propose a bitemporal image transformer (BIT) to efficiently and effectively model contexts within the spatial-temporal domain. Our intuition is that the high-level concepts of the change of interest can be represented by a few visual words, i.e., semantic tokens. To achieve this, we express the bitemporal image into a few tokens, and use a transformer encoder to model contexts in the compact token-based space-time. The learned context-rich tokens are then feedback to the pixel-space for refining the original features via a transformer decoder. We incorporate BIT in a deep feature differencing-based CD framework. Extensive experiments on three CD datasets demonstrate the effectiveness and efficiency of the proposed method. Notably, our BIT-based model significantly outperforms the purely convolutional baseline using only 3 times lower computational costs and model parameters. Based on a naive backbone (ResNet18) without sophisticated structures (e.g., FPN, UNet), our model surpasses several state-of-the-art CD methods, including better than four recent attention-based methods in terms of efficiency and accuracy. Our code is available at this https URL\_CD.
205 citations
TL;DR: A deep translation based change detection network (DTCDN) for optical and SAR images is proposed that utilizes deep context features to separate the unchanged pixels and changed pixels in a supervised CD network.
Abstract: With the development of space-based imaging technology, a larger and larger number of images with different modalities and resolutions are available. The optical images reflect the abundant spectral information and geometric shape of ground objects, whose qualities are degraded easily in poor atmospheric conditions. Although synthetic aperture radar (SAR) images cannot provide the spectral features of the region of interest (ROI), they can capture all-weather and all-time polarization information. In nature, optical and SAR images encapsulate lots of complementary information, which is of great significance for change detection (CD) in poor weather situations. However, due to the difference in imaging mechanisms of optical and SAR images, it is difficult to conduct their CD directly using the traditional difference or ratio algorithms. Most recent CD methods bring image translation to reduce their difference, but the results are obtained by ordinary algebraic methods and threshold segmentation with limited accuracy. Towards this end, this work proposes a deep translation based change detection network (DTCDN) for optical and SAR images. The deep translation firstly maps images from one domain (e.g., optical) to another domain (e.g., SAR) through a cyclic structure into the same feature space. With the similar characteristics after deep translation, they become comparable. Different from most previous researches, the translation results are imported to a supervised CD network that utilizes deep context features to separate the unchanged pixels and changed pixels. In the experiments, the proposed DTCDN was tested on four representative data sets from Gloucester, California, and Shuguang village. Compared with state-of-the-art methods, the effectiveness and robustness of the proposed method were confirmed.
166 citations
TL;DR: The proposed Dual-LSTM framework achieves real-time high-precision RUL Prediction by connecting the change point detection and RUL prediction with the health index construction, introduces a novel one-dimension health index function and leverages historical information to achieve detection and prediction tasks.
145 citations
TL;DR: Wang et al. as discussed by the authors proposed a dual-task constrained deep Siamese convolutional network (DTCDSCN) model, which contains three subnetworks: a change detection network and two semantic segmentation networks.
Abstract: In recent years, building change detection methods have made great progress by introducing deep learning, but they still suffer from the problem of the extracted features not being discriminative enough, resulting in incomplete regions and irregular boundaries. To tackle this problem, we propose a dual-task constrained deep Siamese convolutional network (DTCDSCN) model, which contains three subnetworks: a change detection network and two semantic segmentation networks. DTCDSCN can accomplish both change detection and semantic segmentation at the same time, which can help to learn more discriminative object-level features and obtain a complete change detection map. Furthermore, we introduce a dual attention module (DAM) to exploit the interdependencies between channels and spatial positions, which improves the feature representation. We also improve the focal loss function to suppress the sample imbalance problem. The experimental results obtained with the WHU building data set show that the proposed method is effective for building change detection and achieves state-of-the-art performance in terms of four metrics on the WHU building data set: precision, recall, F1-score, and intersection over union.
129 citations
TL;DR: Deep change vector analysis (DCVA) and fuzzy rules can be applied to identify changed buildings (new/destroyed) in bitemporal SAR images using a cycle-consistent generative adversarial network (CycleGAN).
Abstract: Building change detection (CD), important for its application in urban monitoring, can be performed in near real time by comparing prechange and postchange very-high-spatial-resolution (VHR) synthetic-aperture-radar (SAR) images However, multitemporal VHR SAR images are complex as they show high spatial correlation, prone to shadows, and show an inhomogeneous signature Spatial context needs to be taken into account to effectively detect a change in such images Recently, convolutional-neural-network (CNN)-based transfer learning techniques have shown strong performance for CD in VHR multispectral images However, its direct use for SAR CD is impeded by the absence of labeled SAR data and, thus, pretrained networks To overcome this, we exploit the availability of paired unlabeled SAR and optical images to train for the suboptimal task of transcoding SAR images into optical images using a cycle-consistent generative adversarial network (CycleGAN) The CycleGAN consists of two generator networks: one for transcoding SAR images into the optical image domain and the other for projecting optical images into the SAR image domain After unsupervised training, the generator transcoding SAR images into optical ones is used as a bitemporal deep feature extractor to extract optical-like features from bitemporal SAR images Thus, deep change vector analysis (DCVA) and fuzzy rules can be applied to identify changed buildings (new/destroyed) We validate our method on two data sets made up of pairs of bitemporal VHR SAR images on the city of L’Aquila (Italy) and Trento (Italy)
128 citations
TL;DR: New end-to-end change detection network, called difference-enhancement dense-attention convolutional neural network (DDCNN), that is designed for detection of changes in the bitemporal optical remote sensing images achieves new state-of-the-art change detection performance on these five challenging data sets.
Abstract: This study presents a new end-to-end change detection network, called difference-enhancement dense-attention convolutional neural network (DDCNN), that is designed for detection of changes in the bitemporal optical remote sensing images. To model the internal correlation between high-level and low-level features, a dense attention method consisting of several up-sampling attention units is proposed. Both the up-sampling spatial and up-sampling channel attention are adopted by the unit. The unit, which can use high-level features with rich category information to guide the selection of low-level features, can use the spatial context information to capture the changed features of ground objects. Furthermore, DDCNN also pays attention to the differentiating features of the bitemporal images. By introducing a DE unit, each pixel is weighted and the features are selectively aggregated. The combination of dense attention and the DE unit improves the effectiveness of the network and its accuracy in extracting the change features. The effectiveness of the proposed approach is demonstrated via five challenge data sets. The experimental results show that DDCNN achieves new state-of-the-art change detection performance on these five challenging data sets. For the seasonal change detection data set in particular, compared with the best existing change detection model, the proposed method increases the F1 score and IoU by 2.96% and 5.17%, respectively; compared with the baseline method, our method improved 3.75% and 6.50% on the F1 score and IoU, respectively.
91 citations
TL;DR: This article proposes a novel data-level solution, namely, Instance-level change Augmentation (IAug), to generate bitemporal images that contain changes involving plenty and diverse buildings by leveraging generative adversarial training.
Abstract: Training deep learning-based change detection (CD) models heavily relies on large labeled data sets. However, it is time-consuming and labor-intensive to collect large-scale bitemporal images that contain building change, due to both its rarity and sparsity. Contemporary methods to tackle the data insufficiency mainly focus on transformation-based global image augmentation and cost-sensitive algorithms. In this article, we propose a novel data-level solution, namely, Instance-level change Augmentation (IAug), to generate bitemporal images that contain changes involving plenty and diverse buildings by leveraging generative adversarial training. The key of IAug is to blend synthesized building instances onto appropriate positions of one of the bitemporal images. To achieve this, a building generator is employed to produce realistic building images that are consistent with the given layouts. Diverse styles are later transferred onto the generated images. We further propose context-aware blending for a realistic composite of the building and the background. We augment the existing CD data sets and also design a simple yet effective CD model--CD network (CDNet). Our method (CDNet + IAug) has achieved state-of-the-art results in two building CD data sets (LEVIR-CD and WHU-CD). Interestingly, we achieve comparable results with only 20% of the training data as the current state-of-the-art methods using 100% data. Extensive experiments have validated the effectiveness of the proposed IAug. Our augmented data set has a lower risk of class imbalance than the original one. Conventional learning on the synthesized data set outperforms several popular cost-sensitive algorithms on the original data set. Our code and data are available at https://github.com/justchenhao/IAug_CDNet.
TL;DR: A method based on a multiscales fully convolutional neural network (MFCN), which uses multiscale convolution kernels to extract the detailed features of the ground object features and a loss function combining weighted binary cross-entropy (WBCE) loss and dice coefficient loss is proposed, so that the model can be trained from unbalanced samples.
Abstract: In the task of change detection (CD), high-resolution remote sensing images (HRSIs) can provide rich ground object information. However, the interference from noise and complex background information can also bring some challenges to CD. In recent years, deep learning methods represented by convolutional neural networks (CNNs) have achieved good CD results. However, the existing methods have difficulty in detecting the detailed change information of the ground objects effectively. The imbalance of positive and negative samples can also seriously affect the CD results. In this letter, to solve the above problems, we propose a method based on a multiscale fully convolutional neural network (MFCN), which uses multiscale convolution kernels to extract the detailed features of the ground object features. A loss function combining weighted binary cross-entropy (WBCE) loss and dice coefficient loss is also proposed, so that the model can be trained from unbalanced samples. The proposed method was compared with six state-of-the-art CD methods on the DigitalGlobe dataset. The experiments showed that the proposed method can achieve a higher F1-score, and the detection effect of the detailed changes was better than that of the other methods.
TL;DR: A new change detection method based on similarity measurement between heterogeneous images that can avoid the leakage of heterogeneous data and bring more robust change detection results is proposed.
TL;DR: The post-classification change detection method using maximum likelihood classifier (MLC) supervised classification is applicable in all cases and is the most commonly used technique from the past till present that has achieved high accuracy in all regions comparatively to other techniques.
TL;DR: A novel Cross Layer convolutional neural Network (CLNet) is proposed, where the UNet structure is used as the backbone and newly designed Cross Layer Blocks (CLBs) are embedded to incorporate the multi-scale features and multi-level context information.
Abstract: Change detection plays a crucial role in observing earth surface transition and has been widely investigated using deep learning methods. However, the current deep learning methods for pixel-wise change detection still suffer from limited accuracy, mainly due to their insufficient feature extraction and context aggregation. To address this limitation, we propose a novel Cross Layer convolutional neural Network (CLNet) in this paper, where the UNet structure is used as the backbone and newly designed Cross Layer Blocks (CLBs) are embedded to incorporate the multi-scale features and multi-level context information. The designed CLB starts with one input and then split into two parallel but asymmetric branches, which are leveraged to extract the multi-scale features by using different strides; and the feature maps, which come from the opposite branches but have the same size, are concatenated to incorporate multi-level context information. The designed CLBs aggregate the multi-scale features and multi-level context information so that the proposed CLNet can reuse extracted feature information and capture accurate pixel-wise change in complex scenes. Quantitative and qualitative experiments were conducted on a public very-high-resolution satellite image dataset (VHR-Dataset), a newly released building change detection dataset (LEVIR-CD Dataset) and an aerial building change detection dataset (WHU Building Dataset). The CLNet reached an F1-score of 0.921 and an overall accuracy of 98.1% with the VHR-Dataset, an F1-score of 0.900 and an overall accuracy of 98.9% with the LEVIR-CD Dataset, and an F1-score of 0.963 and an overall accuracy of 99.7% with the WHU Building Dataset. The experimental results with all the selected datasets showed that the proposed CLNet outperformed several state-of-the-art (SOTA) methods and achieved competitive accuracy and efficiency trade-offs. The code of CLNet will be released soon at: https://skyearth.org/publication/project/CLNet .
TL;DR: Wang et al. as discussed by the authors proposed a super-resolution-based change detection network (SRCDNet), which employs a super resolution (SR) module containing a generator and a discriminator to directly learn SR images through adversarial learning and overcome the resolution difference between bi-temporal images.
Abstract: Change detection, which aims to distinguish surface changes based on bi-temporal images, plays a vital role in ecological protection and urban planning. Since high resolution (HR) images cannot be typically acquired continuously over time, bi-temporal images with different resolutions are often adopted for change detection in practical applications. Traditional subpixel-based methods for change detection using images with different resolutions may lead to substantial error accumulation when HR images are employed; this is because of intraclass heterogeneity and interclass similarity. Therefore, it is necessary to develop a novel method for change detection using images with different resolutions, that is more suitable for HR images. To this end, we propose a super-resolution-based change detection network (SRCDNet) with a stacked attention module. The SRCDNet employs a super resolution (SR) module containing a generator and a discriminator to directly learn SR images through adversarial learning and overcome the resolution difference between bi-temporal images. To enhance the useful information in multi-scale features, a stacked attention module consisting of five convolutional block attention modules (CBAMs) is integrated to the feature extractor. The final change map is obtained through a metric learning-based change decision module, wherein a distance map between bi-temporal features is calculated. The experimental results demonstrate the superiority of the proposed method, which not only outperforms all baselines -with the highest F1 scores of 87.40% on the building change detection dataset and 92.94% on the change detection dataset -but also obtains the best accuracies on experiments performed with images having a 4x and 8x resolution difference. The source code of SRCDNet will be available at this https URL.
13 Apr 2021
TL;DR: Some new dimensions that emerge at the intersection of sequential change detection with other areas are discussed, along with a selection of modern applications and remarks on open questions.
Abstract: Online detection of changes in stochastic systems, referred to as sequential change detection or quickest change detection, is an important research topic in statistics, signal processing, and information theory, and has a wide range of applications. This survey starts with the basics of sequential change detection, and then moves on to generalizations and extensions of sequential change detection theory and methods. We also discuss some new dimensions that emerge at the intersection of sequential change detection with other areas, along with a selection of modern applications and remarks on open questions.
TL;DR: The results of quantitative analysis and qualitative comparison indicate that the ADS-Net method comprises better effectiveness and robustness compared to the other state-of-the-art change detection methods.
TL;DR: A novel triplet input network is introduced, which is capable of learning bi-temporal image features, extracting the temporal information reflecting the difference between images over time, and the effectiveness and robustness of HRTNet are verified on three popular high-resolution remote sensing image datasets.
Abstract: Change detection in remote sensing images aims to accurately determine any significant land surface changes based on acquired multi-temporal image data, being a pivotal task of remote sensing image processing. Over the past few years, owing to its powerful learning and expression ability, deep learning has been widely applied in the general field of image processing and has demonstrated remarkable potentials in performing change detection in images. However, a majority of the existing deep learning-based change detection mechanisms are modified from single-image semantic segmentation algorithms, without considering the temporal information contained within the images, thereby not always appropriate for real-world change detection. This paper proposes a High-Resolution Triplet Network (HRTNet) framework, including a dynamic inception module, to tackle such shortcomings in change detection. First, a novel triplet input network is introduced, which is capable of learning bi-temporal image features, extracting the temporal information reflecting the difference between images over time. Then, a network is employed to extract high-resolution image features, ensuring the learned features preserving high-resolution characteristics with minimal reduction of information. The paper also proposes a novel dynamic inception module, which helps improve the feature expression ability of HRTNet, enriching the multi-scale information of the features extracted. Finally, the distances between feature pairs are measured to generate a high-precision change map. The effectiveness and robustness of HRTNet are verified on three popular high-resolution remote sensing image datasets. Systematic experimental results show that the proposed approach outperforms state-of-the-art change detection methods.
TL;DR: This article aims to provide an empirical review of the state-of-the-art deep learning methods for change detection, providing a detailed analysis of the technical characteristics of different model designs and experimental frameworks and provides model design based categorization of the existing approaches.
Abstract: Visual change detection, aiming at segmentation of video frames into foreground and background regions, is one of the elementary tasks in computer vision and video analytics. The applications of change detection include anomaly detection, object tracking, traffic monitoring, human machine interaction, behavior analysis, action recognition, and visual surveillance. Some of the challenges in change detection include background fluctuations, illumination variation, weather changes, intermittent object motion, shadow, fast/slow object motion, camera motion, heterogeneous object shapes and real-time processing. Traditionally, this problem has been solved using hand-crafted features and background modelling techniques. In recent years, deep learning frameworks have been successfully adopted for robust change detection. This article aims to provide an empirical review of the state-of-the-art deep learning methods for change detection. More specifically, we present a detailed analysis of the technical characteristics of different model designs and experimental frameworks. We provide model design based categorization of the existing approaches, including the 2D-CNN, 3D-CNN, ConvLSTM, multi-scale features, residual connections, autoencoders and GAN based methods. Moreover, an empirical analysis of the evaluation settings adopted by the existing deep learning methods is presented. To the best of our knowledge, this is a first attempt to comparatively analyze the different evaluation frameworks used in the existing deep change detection methods. Finally, we point out the research needs, future directions and draw our own conclusions.
TL;DR: This work proposes a computationally simple novel algorithm for network change point localization, called Network Binary Segmentation, which relies on weighted averages of the adjacency matrices, and devise a more sophisticated algorithm based on singular value thresholding, called Local Refinement, that delivers more accurate estimates of the change point locations.
Abstract: We study the problem of change point detection and localization in dynamic networks We assume that we observe a sequence of independent adjacency matrices of given size, each corresponding to one realization from an unknown inhomogeneous Bernoulli model The underlying distribution of the adjacency matrices may change over a subset of the time points, called change points Our task is to recover with high accuracy the unknown number and positions of the change points Our generic model setting allows for all the model parameters to change with the total number of time points, including the network size, the minimal spacing between consecutive change points, the magnitude of the smallest change and the degree of sparsity of the networks We first identify an impossible region in the space of the model parameters such that no change point estimator is provably consistent if the data are generated according to parameters falling in that region We propose a computationally simple novel algorithm for network change point localization, called Network Binary Segmentation, which relies on weighted averages of the adjacency matrices We show that Network Binary Segmentation is consistent over a range of the model parameters that nearly cover the complement of the impossibility region, thus demonstrating the existence of a phase transition for the problem at hand Next, we devise a more sophisticated algorithm based on singular value thresholding, called Local Refinement, that delivers more accurate estimates of the change point locations We show that, under appropriate conditions, Local Refinement guarantees a minimax optimal rate for network change point localization while remaining computationally feasible
TL;DR: A semisupervised CD method that encodes mult itemporal images as a graph via multiscale parcel segmentation that effectively captures the spatial and spectral aspects of the multitemporal images.
Abstract: Most change detection (CD) methods are unsupervised as collecting substantial multitemporal training data is challenging. Unsupervised CD methods are driven by heuristics and lack the capability to learn from data. However, in many real-world applications, it is possible to collect a small amount of labeled data scattered across the analyzed scene. Such a few scattered labeled samples in the pool of unlabeled samples can be effectively handled by graph convolutional network (GCN) that has recently shown good performance in semisupervised single-date analysis, to improve change detection performance. Based on this, we propose a semisupervised CD method that encodes multitemporal images as a graph via multiscale parcel segmentation that effectively captures the spatial and spectral aspects of the multitemporal images. The graph is further processed through GCN to learn a multitemporal model. Information from the labeled parcels is propagated to the unlabeled ones over training iterations. By exploiting the homogeneity of the parcels, the model is used to infer the label at a pixel level. To show the effectiveness of the proposed method, we tested it on a multitemporal Very High spatial Resolution (VHR) data set acquired by Pleiades sensor over Trento, Italy.
TL;DR: In this article, a comprehensive review of change detection in very high-spatial-resolution (VHR) images is presented, which mainly includes three aspects: methods, applications, and future directions.
Abstract: Change detection is a vibrant area of research in remote sensing. Thanks to increases in the spatial resolution of remote sensing images, subtle changes at a finer geometrical scale can now be effectively detected. However, change detection from very-high-spatial-resolution (VHR) (≤5 m) remote sensing images is challenging due to limited spectral information, spectral variability, geometric distortion, and information loss. To address these challenges, many change detection algorithms have been developed. However, a comprehensive review of change detection in VHR images is lacking in the existing literature. This review aims to fill the gap and mainly includes three aspects: methods, applications, and future directions.
TL;DR: A novel integrated approach combining a deep convolutional neural network (CNN) and change detection is proposed for landslide recognition from RS images, which has an accuracy exceeding 80%, and the experiments demonstrate its high practicability.
Abstract: It is a technological challenge to recognize landslides from remotely sensed (RS) images automatically and at high speeds, which is fundamentally important for preventing and controlling natural landslide hazards. Many methods have been developed, but there remains room for improvement for stable, higher accuracy, and high-speed landslide recognition for large areas with complex land cover. In this article, a novel integrated approach combining a deep convolutional neural network (CNN) and change detection is proposed for landslide recognition from RS images. Logically, it comprises the following four parts. First, a CNN for landslide recognition is built based on training data sets from RS images with historical landslides. Second, the object-oriented change detection CNN (CDCNN) with a fully connected conditional random field (CRF) is implemented based on the trained CNN. Third, the preliminary CDCNN is optimized by the proposed postprocessing methods. Finally, the results are further enhanced by a set of information extraction methods, including trail extraction, source point extraction, and attribute extraction. Furthermore, in the implementation of the proposed approach, image block processing and parallel processing strategies are adopted. As a result, the speed has been improved significantly, which is extremely important for RS images covering large areas. The effectiveness of the proposed approach has been examined using two landslide-prone sites, Lantau Island and Sharp Peak, Hong Kong, with a total area of more than 70 km2. Besides its high speed, the proposed approach has an accuracy exceeding 80%, and the experiments demonstrate its high practicability.
TL;DR: The potential impact of automation to decision making is demonstrated by deploying a novel change detection method implementing change feature extraction using convolutional neural networks under an OBIA framework to map a large geographic area affected by a recent natural disaster.
TL;DR: A novel unsupervised deep-learning-based CD method that can effectively model contextual information and handle the large number of bands in multispectral HR images is presented.
Abstract: To overcome the limited capability of most state-of-the-art change detection (CD) methods in modeling spatial context of multispectral high spatial resolution (HR) images and exploiting all spectral bands jointly, this letter presents a novel unsupervised deep-learning-based CD method that can effectively model contextual information and handle the large number of bands in multispectral HR images. This is achieved by exploiting all spectral bands after grouping them into spectral-dedicated band groups. To eliminate the necessity of multitemporal training data, the proposed method exploits a data set targeted for image classification to train spectral-dedicated Auxiliary Classifier Generative Adversarial Networks (ACGANs). They are used to obtain pixelwise deep change hypervector from multitemporal images. Each feature in deep change hypervector is analyzed based on the magnitude to identify changed pixels. An ensemble decision fusion strategy is used to combine change information from different features. Experimental results on the urban, Alpine, and agricultural Sentinel-2 data sets confirm the effectiveness of the proposed method.
TL;DR: A convolutional neural network with a multi-objective cost function taking care of spatial and statistical properties of the SAR image is proposed, achieved by the definition of a peculiar loss function obtained by the weighted combination of three different terms.
Abstract: Deep learning (DL) in remote sensing has nowadays become an effective operative tool: it is largely used in applications, such as change detection, image restoration, segmentation, detection, and classification. With reference to the synthetic aperture radar (SAR) domain, the application of DL techniques is not straightforward due to the nontrivial interpretation of SAR images, especially caused by the presence of speckle. Several DL solutions for SAR despeckling have been proposed in the last few years. Most of these solutions focus on the definition of different network architectures with similar cost functions, not involving SAR image properties. In this article, a convolutional neural network (CNN) with a multi-objective cost function taking care of spatial and statistical properties of the SAR image is proposed. This is achieved by the definition of a peculiar loss function obtained by the weighted combination of three different terms. Each of these terms is dedicated mainly to one of the following SAR image characteristics: spatial details, speckle statistical properties, and strong scatterers identification. Their combination allows balancing these effects. Moreover, a specifically designed architecture is proposed to effectively extract distinctive features within the considered framework. Experiments on simulated and real SAR images show the accuracy of the proposed method compared with the state-of-art despeckling algorithms, both from a quantitative and qualitative point of view. The importance of considering such SAR properties in the cost function is crucial for correct noise rejection and details preservation in different underlined scenarios, such as homogeneous, heterogeneous, and extremely heterogeneous.
TL;DR: A cross-resolution difference learning method is proposed to detect changes from multitemporal images in the originally different resolutions without resizing operations and demonstrates the effectiveness of this method for detecting changes from different resolution images.
Abstract: Change detection (CD) aims to identify the differences between multitemporal images acquired over the same geographical area at different times. With the advantages of requiring no cumbersome labeled change information, unsupervised CD has attracted extensive attention of researchers. Multitemporal images tend to have different resolutions as they are usually captured at different times with different sensor properties. It is difficult to directly obtain one pixelwise change map for two images with different resolutions, so current methods usually resize multitemporal images to a unified size. However, resizing operations change the original information of pixels, which limits the final CD performance. This article aims to detect changes from multitemporal images in the originally different resolutions without resizing operations. To achieve this, a cross-resolution difference learning method is proposed. Specifically, two cross-resolution pixelwise difference maps are generated for the two different resolution images and fused to produce the final change map. First, the two input images are segmented into individual homogeneous regions separately due to different resolutions. Second, each pixelwise difference map is produced according to two measure distances, the mutual information distance and the deep feature distance, between image regions in which the pixel lies. Third, the final binary change map is generated by fusing and binarizing the two cross-resolution difference maps. Extensive experiments on four datasets demonstrate the effectiveness of the proposed method for detecting changes from different resolution images.
TL;DR: A review of recent developments, opportunities, and trends in Landsat change detection studies can be found in this paper, where a representative database was created resulting in 490 relevant journal articles derived from the Web of Science and Scopus.
Abstract: With uninterrupted space-based data collection since 1972, Landsat plays a key role in systematic monitoring of the Earth’s surface, enabled by an extensive and free, radiometrically consistent, global archive of imagery. Governments and international organizations rely on Landsat time series for monitoring and deriving a systematic understanding of the dynamics of the Earth’s surface at a spatial scale relevant to management, scientific inquiry, and policy development. In this study, we identify trends in Landsat-informed change detection studies by surveying 50 years of published applications, processing, and change detection methods. Specifically, a representative database was created resulting in 490 relevant journal articles derived from the Web of Science and Scopus. From these articles, we provide a review of recent developments, opportunities, and trends in Landsat change detection studies. The impact of the Landsat free and open data policy in 2008 is evident in the literature as a turning point in the number and nature of change detection studies. Based upon the search terms used and articles included, average number of Landsat images used in studies increased from 10 images before 2008 to 100,000 images in 2020. The 2008 opening of the Landsat archive resulted in a marked increase in the number of images used per study, typically providing the basis for the other trends in evidence. These key trends include an increase in automated processing, use of analysis-ready data (especially those with atmospheric correction), and use of cloud computing platforms, all over increasing large areas. The nature of change methods has evolved from representative bi-temporal pairs to time series of images capturing dynamics and trends, capable of revealing both gradual and abrupt changes. The result also revealed a greater use of nonparametric classifiers for Landsat change detection analysis. Landsat-9, to be launched in September 2021, in combination with the continued operation of Landsat-8 and integration with Sentinel-2, enhances opportunities for improved monitoring of change over increasingly larger areas with greater intra- and interannual frequency.