In recent years, change detection (CD) has achieved remarkable success through using deep learning. However, most existing methods rely on label teaching, and thus have many limitations when dealing with the complexity and diversity of remote sensing scenes. In this work, we propose a self-supervised learning pretraining framework for remote sensing image CD (SSLCD). Our motivation is to leverage the intrinsic structure of multitemporal data to learn general and robust representations, generating competitive pretrained models for the CD task. On the one hand, an intrinsic structure learning strategy is introduced, which enforces feature invariance and change consistency across temporal phases and augmented views, learning discriminatory representations related to changes while simultaneously mitigating noises associated with irrelevant changes. On the other hand, a self/cross-reconstruction mechanism is proposed, which extends masked image modeling to multitemporal images by predicting missing parts of the pre-phase image using the post-phase image, thereby enhancing the model’s capacity in modeling high-level contextual information. Finally, models pretrained using SSLCD are extensively evaluated on three CD datasets, and the results demonstrate that SSLCD outperforms existing remote sensing pretraining methods as well as the state-of-the-art CD methods.

API plays an important role in modern software development. Automatic API recommendation has been studied for years to facilitate developers' learning process of APIs. Previous approaches mainly use statistical models and collab-orative filtering (CF) techniques to mine API usage patterns for recommendation. Despite the encouraging results, they still struggle to obtain the accurate embeddings of the client methods and called APIs. Prior studies generally formulate the process of API call interactions as undirected graph structure, neglecting the order in which the API invocations appear, thus fail to seize the rich relationship and complex transitions of API calls. To transcend the limitations, we propose a novel method, namely PARO, to predict the next API invocations using gated graph neural networks (GGNNs). In our proposed method, the API call sequences are modeled as directed graphs, thus the GNN models prone to capture features such as the partial order and complex transitions between API invocations. Besides, we also learn the text attribute representations of API invocations and client methods through word embedding, which further corroborates the semantic and lexical similarities between them. We conduct experimental evaluations on a large number of Java projects extracted from Github and Maven Central. Results show that our approach outperforms the state-of-the-art by a large margin, in terms of Hit@N and MRR@N.

Existing change detection methods mainly focus on bitemporal remote sensing images with a wide time range, which contain several changed areas. However, short-term change detection, which covers a smaller time range, has lower temporal resolution and more unbalanced positive and negative samples, leading to performance degradation. To solve this issue, we propose a multitask framework named Time-Correlated and Edge Information-Guided Network for Short-Term Building Change Detection (TCEIGNet). A time-correlated block is introduced to link the current image with historical information, enhancing the model’s sensitivity to time information. To address the blurring of prediction boundaries and the lack of local details, we propose to use a difference-of-Gaussian (DoG) pyramid to incorporate boundary prior knowledge at each level. Additionally, we add a feature constraint loss to tackle sample imbalance. Experimental results on SpaceNet7 demonstrate that our method outperforms other network methods at different time intervals.

UMRS can effectively explore the aquifer information in front of the tunnel face. However, the complex tunnel construction environment will reduce the reliability of the interpretation results. We propose and implement using joint inversion of UMRS-TEM data in tunnel detection using HMC for the first time to solve this problem. Joint inversion can update the UMRS kernel in real-time during the iterative process, obtain accurate aquifer information and resistivity structure, and improve the reliability of the interpretation of inversion. HMC is a MCMC method that uses Hamiltonian dynamics to propose future states in Markov chains. It can explore the target distribution more effectively, and it also has the advantage that MCMC can obtain a posteriori PDF of parameters, we believe that PDF information can effectively judge the accuracy of inversion results. We validated the effectiveness and practicality of the joint inversion using synthetic and observed data, and the experimental results showed its advantages in accuracy and noise resistance compared to a single UMRS inversion. In the field example, the water content PDF of the joint inversion is significantly increased by 31.8% compared with that of the single inversion. We analyzed the correlation between water content, resistivity, and layer interfaces and discovered some correlation laws. HMC improves the efficiency of computational and provides assistance for further research on the influence of parameters on inversion results. Our conclusions can improve the safety of tunnel construction and provide effective technical support for avoiding hydrogeological disasters in tunnels.

Underground magnetic resonance sounding (UMRS) has the problem of low data quantity and low data quality in tunnel detection, a probabilistic statistical method is needed for data inversion. We used Hamiltonian Monte Carlo (HMC) to obtain UMRS inversion results, and we implemented a “tempered” scheme in HMC, in order to obtain higher efficiency and accuracy of inversion. This is the first time tempered HMC (THMC) has been applied to UMRS inversion and tunnel detection. It adds the neglected temperature term into HMC, effectively improving the escape ability, and improving computational efficiency. First, UMRS and THMC methods are briefly introduced in this article. Then, we investigate the relationship between different temperatures and the ability of THMC to jump out of the local optimal and find the temperature range suitable for UMRS inversion. We designed a series of schemes to test the performance of two methods and demonstrate that UMRS inversion using THMC has clear advantages. The inversion results of synthetic data show that THMC has higher efficiency and accuracy than HMC under extreme conditions, such as weak signal and high noise. Finally, we introduce the general situation of the study site and apply the two methods to the observation data inversion. THMC obtains results that are more consistent with the actual situation, which proves that it has strong practicability. We believe that THMC is more suitable for UMRS data inversion than HMC. THMC is helpful in improving the detection accuracy and efficiency of UMRS, ensuring the safety of tunnel construction, and preventing the delay of the construction period.

Surface magnetic resonance sounding (SMRS) is a unique geophysical method that can directly track and quantify groundwater using the remote sensing technique. The conventional SMRS used a closed coil as the magnetic field source. In the field measurement, a large-size coil is laid on the ground or even multiple coils are set for array detection. This reduces the detection efficiency and consumes labor inevitably. The current study proposes an electrical source (ES), a new mode for exciting the groundwater. It is a long wire placed on the ground and connected to the Earth by two grounding electrodes. The ES has the advantages of labor-saving, time-saving, and better environmental adaptability. Moreover, the magnetic field generated by the ES can transmit farther than the traditional magnetic source. Based on this, we matched different receivers for the ES and simulated the kernel, the resolution, and the signal with different configurations. The results show that using the long grounding wire as both the electrical transmitter and receiver can obtain higher signal amplitude and better resolution than the traditional magnetic configuration. In addition, it has the ability to break through the detection depth of the conventional method.

Deep learning-based change detection methods have achieved remarkable success through the feature learning capability of deep convolutions. However, the network structures of existing methods are simply modified from the semantic segmentation models, ignoring the essential characteristics of change detection, thereby limiting their applications. In this work, we propose a category-awareness based difference-threshold alternative learning network (D-TNet) for remote sensing image change detection. Our motivation is to characterize the different change magnitudes for different land cover changes, and represent the semantic content differences of various objects. Thus, our D-TNet consists of a difference map learning path and a threshold map learning path, realizing self-adapting thresholds selection by assigning each pixel a unique threshold. The two paths are alternatively optimized to make the difference map more discriminative, as well as making the threshold map more adaptive. In addition, a category-awareness attention mechanism is introduced in D-TNet, which learns a pixel-to-category relationship to benefit in representing the heterogeneity of land covers. Finally, experimental results on three change detection datasets verify the effectiveness of our D-TNet in both visual and quantitative analysis. Code will be available at: https://www.researchgate.net/profile/Ling-Wan-4.

Change detection for high resolution remote sensing images is an important but challenging task. In this article, we propose a spatial-temporal-channel attention Unet++ (STC-Unet++) for remote sensing image change detection. The STC-Unet++ takes advantage of the Unet++ structure, combining semantic information to change detection. In addition, it employs a spatial-temporal-channel attention mechanism, extracting features more discriminatively and improving the change detection accuracy without increasing training time. Finally, experiments are carried out on the LEVIR-CD dataset, and the results show that the STC-Unet++ can effectively detect the changes, achieving 89.0% recall, 88.3% accuracy, 88.4% F1-score, 79.49% IoU and 94.1% AUC.

Change detection in SAR images is an important but challenge task. Due to the difficulty of SAR interpretation, reliable training samples are lacking, limiting the application of deep learning technology in SAR image change detection. To overcome this problem, this article proposes an unsupervised SAR image change detection method based on slow feature analysis theory with convolutional neural network (SAR-SFAnet). It adopts SDAEs to automatically extract features from SAR data, and employs slow feature analysis theory to project the extracted multi -dimensional features into a new space. In addition, an alternative optimization strategy is introduced, making the features learned by bi - temporal stacked denoising auto-encoder (SDAEs) have more consistent representations, as well as making the change detection map more accurate. Finally, comparative experiments are carried out on two real SAR data sets, demonstrating the effectiveness of the proposed method.

BACKGROUND: The 10th and 9th revisions of the International Statistical Classification of Diseases and Related Health Problems (ICD10 and ICD9) have been adopted worldwide as a well-recognized norm to share codes for diseases, signs and symptoms, abnormal findings, etc. The international Consortium for Clinical Characterization of COVID-19 by EHR (4CE) website stores diagnosis COVID-19 disease data using ICD10 and ICD9 codes. However, the ICD systems are difficult to decode due to their many shortcomings, which can be addressed using ontology. METHODS: An ICD ontology (ICDO) was developed to logically and scientifically represent ICD terms and their relations among different ICD terms. ICDO is also aligned with the Basic Formal Ontology (BFO) and reuses terms from existing ontologies. As a use case, the ICD10 and ICD9 diagnosis data from the 4CE website were extracted, mapped to ICDO, and analyzed using ICDO. RESULTS: We have developed the ICDO to ontologize the ICD terms and relations. Different from existing disease ontologies, all ICD diseases in ICDO are defined as disease processes to describe their occurrence with other properties. The ICDO decomposes each disease term into different components, including anatomic entities, process profiles, etiological causes, output phenotype, etc. Over 900 ICD terms have been represented in ICDO. Many ICDO terms are presented in both English and Chinese. The ICD10/ICD9-based diagnosis data of over 27,000 COVID-19 patients from 5 countries were extracted from the 4CE. A total of 917 COVID-19-related disease codes, each of which were associated with 1 or more cases in the 4CE dataset, were mapped to ICDO and further analyzed using the ICDO logical annotations. Our study showed that COVID-19 targeted multiple systems and organs such as the lung, heart, and kidney. Different acute and chronic kidney phenotypes were identified. Some kidney diseases appeared to result from other diseases, such as diabetes. Some of the findings could only be easily found using ICDO instead of ICD9/10. CONCLUSIONS: ICDO was developed to ontologize ICD10/10 codes and applied to study COVID-19 patient diagnosis data. Our findings showed that ICDO provides a semantic platform for more accurate detection of disease profiles.

Phase correlation (PC), an efficient frequency-domain registration method, has been extensively used in remote sensing images owing to its subpixel accuracy and robustness to image contrast, noise, and occlusions. However, its performance becomes poor when applied to the registration between optical and synthetic aperture radar (SAR) images, which are two typical multisensor images. Inspired by the recently proposed feature-based methods, we present a novel subpixel registration method that combines robust feature representations of optical and SAR images and the 3-D PC (OS-PC). The robust feature representations, which capture the inherent property of the two images and retain their structural information, form two dense image cubes. The 3-D PC utilizes the image cubes as a substitute of two raw images to estimate 2-D translations, either by locating peak in the spatial domain or by directly working in the Fourier domain. Furthermore, we investigate two techniques to improve the accuracy of the 3-D PC both in the spatial domain and Fourier domain: the first is the constrained energy minimization method to seek the Dirac delta function after 3-D inverse Fourier transform and the second is the fast sample consensus fitting to estimate phase difference after high-order singular value decomposition of the PC matrix. Experiments with both simulated and satellite optical-to-SAR pairs were carried out to test the proposed method. Compared with state-of-the-art PC methods and optical-to-SAR registration methods, the proposed method presents a superior performance in both accuracy and robustness. Moreover, we verify the adaptability of the proposed method.

This paper proposes an extension version of our previous work MS-CC to achieve optical and SAR images change detection. The proposed method introduces a cooperative multitemporal segmentation, whose merging process considers the heterogeneity of SAR and optical images as parallel information, making sure that the multitemporal information can be fully utilized without interfering with each other. Then, the change detection strategy based on compound classification is carried out on the segmentation results, obtaining the multi-scale change detection maps. Experimental validation is conducted with GoaFen3 and Google Earth data.

This letter proposes a method for the change detection in multisensor remote sensing images. The proposed method combines multitemporal segmentation and compound classification. In consideration of the particularity of multisensor images, multitemporal segmentation is applied to generate homogeneous objects. This process can reduce the salt and pepper effect that is inevitable in pixel-based methods and reduce the false alarms caused by area transitions and object misalignment in traditional object-based methods. Then, compound classification is carried out at the object level. This process exploits temporal correlations and overcomes the error propagation of traditional postclassification comparison methods. The change map is generated by comparing the classification maps at different times. Experimental validation is conducted with GaoFen3, Terrasar, GaoFen2, and Google Earth data.

During the excavation of underground tunnels and in ore mining, accidents related to water bursts occasionally occur. As the only technique used for the direct detection of groundwater, the nuclear magnetic resonant (NMR) method has advantages for the detection of disaster-inducing water flows. Unfortunately, the amplitudes of underground NMR (UNMR) signals are in the range of some tens of nanovolts (10-9V) or even picovolts (10-12V), and thus extremely susceptible to environmental noise. By increasing the macromagnetic moment of groundwater, both adiabatic pulses and prepolarization (PP) methods have been employed in surface NMR. However, when using either method, it is difficult to achieve substantial signal enhancements over large volumes. For maximum signal amplitudes, we integrated these two approaches and derived the forward formulas with adiabatic pulses under PP for UNMR. In comparison with existing methods, this new model can achieve high sensitivity and broad responses. (A 6-m antenna attains a 10-5V signal level for a homogeneous subsurface with 0.2 m3/m3water content.) Thus, better resolution could also be provided even in a high-noise place. Overall, the large NMR signals and high resolutions make the combination of adiabatic pulses with PP a valuable approach, which is expected to open up a new application for UNMR.

Change detection in heterogeneous remote sensing images is an important but challenging task because of the incommensurable appearances of the heterogeneous images. In order to solve the change detection problem in optical and synthetic aperture radar (SAR) images, this paper proposes an improved method that combines cooperative multitemporal segmentation and hierarchical compound classification (CMS-HCC) based on our previous work. Considering the large radiometric and geometric differences between heterogeneous images, first, a cooperative multitemporal segmentation method is introduced to generate multi-scale segmentation results. This method segments two images together by associating the information from the two images and thus reduces the noises and errors caused by area transition and object misalignment, as well as makes the boundaries of detected objects described more accurately. Then, a region-based multitemporal hierarchical Markov random field (RMH-MRF) model is defined to combine spatial, temporal, and multi-level information. With the RMH-MRF model, a hierarchical compound classification method is performed by identifying the optimal configuration of labels with a region-based marginal posterior mode estimation, further improving the change detection accuracy. The changes can be determined if the labels assigned to each pair of parcels are different, obtaining multi-scale change maps. Experimental validation is conducted on several pairs of optical and SAR images. It consists of two parts: comparison on different multitemporal segmentation methods and comparison on different change detection methods. The results show that the proposed method can effectively detect the changes in heterogeneous images, with low false positive and high accuracy.

Coregistration of high-resolution optical and synthetic aperture radar (SAR) images is still an ongoing problem due to different imaging mechanisms of two kinds of remote sensing images. In this paper, we propose an optical flow-based algorithm to solve the dense registration problem [optical-to-SAR (OS)-flow]. Unlike parametric registration methods that estimate a transformation model, OS-flow aims to find pixelwise correspondences between optical and SAR images. Specifically, two frameworks of OS-flow, a global method and a local method, are proposed. Due to the drastic differences between SAR and optical images, two dense feature descriptors, rather than the raw intensities, are utilized to retain the constancy assumption in optical flow estimation. Considering the inherent properties of the two images, two dense descriptors are constructed using consistent gradient computation. After satisfying the constancy assumption, the global method estimates the flow map by optimizing an objective function, and the local method iteratively estimates the flow vector in a local neighborhood. Both methods use the coarse-to-fine matching strategy to address large displacements and reduce the computational cost. Experiments on several optical-to-SAR image pairs in various scenarios show that the proposed methods have a strong ability to match across optical and SAR images and outperform other state-of-the-art methods in terms of registration accuracy.

The ratio of averages is a robust edge detector which provides the property of constant false alarm rate for synthetic aperture radar (SAR) imagery. However, the rectangular window used in the calculation of local mean may cause numerous false maxima. The size of the processing window also has a significant effect on the detection performance, but it is difficult to determine the optimum window size. In this letter, we first propose a new ratio-based detector that is constructed by the Gabor odd filter. The scale of the proposed detector is related to the size of the processing window. Then, edge strength maps extracted by multiscale detectors are combined using an edge tracking algorithm to form a final response. We used the receiver operating characteristic curves to evaluate the performance of the proposed detector. The experimental results on simulated and real-world SAR images show that the proposed multiscale edge detector yields an accurate and consecutive edge response.