Search

Earthquake Triggered Landslides (ETLs) are serious secondary hazards of earthquakes, causing severe casualties and property losses, and their rapid, automated and accurate detection is of great value. Due to the all-day and all-weather imaging capability of Synthetic Aperture Radar (SAR), ETL detection using SAR images is promising but faces the problem of insufficient accuracy. In this study, we proposed a deep learning-based ETL detection method to address this problem. Firstly, published ETL inventories and SAR images are combined to generate high-quality training datasets. Then, a landslide detection network of Multi-level Features Effective Weighting and Fusion (MFEWF) is proposed to effectively extract and fuze multi-level landslide features to identify SAR pixels within ETLs. Finally, the ETL boundary is determined based on these identified SAR pixels. This method is verified through three earthquake cases: the 2017 Mainling, China earthquake, the 2018 Palu, Indonesia earthquake and the 2018 Papua New Guinea earthquake. Results show that our method can effectively identify landslide boundaries with high accuracy (88.8%, 81.4% and 82.4% for the three cases), obviously outperforming other deep learning frameworks (e.g. DeepLabV3+). Using Sentinel-1 imagery to achieve such high accuracy in landslide detection, this study will improve emergency response to landslide disasters following earthquakes.

Attention mechanisms have found extensive application in Deep Neural Networks (DNNs), with numerous experiments over time showcasing their efficacy in improving the overall performance of DNNs. Considering the black-box nature of DNNs, it is unclear how attention mechanisms affect model decision-making processes. Therefore, it is of great significance to explore the internal relationship between the focused area of attention mechanisms and the prediction results. For the first time, an explainable framework for the attention mechanism of Synthetic Aperture Radar (SAR) image analytics is proposed, integrating DNN and eXplainable Artificial Intelligence (XAI), which realizes the effective explanation for attention mechanisms across different locations. The framework consists of three parts: the water extraction network which contains the attention mechanism at different locations; the proposed Attention- Gradient Class Activation Mapping (A-GCAM) method for attribution analysis of attention mechanisms; and we invent Category-specific channel Score of Confidence Mapping (CSCM) to perform geo-visualization for the output features of attention mechanisms. Experiments are conducted with the widely used Sentinel-1 system for water detection, in which three attention mechanisms with different characteristics are added to the encoder and decoder in DNNs. The results show that the framework can make the decision-making process of attention mechanisms transparent, thus improving their comprehensiveness and trustworthiness in various tasks and providing a reliable approach to selecting suitable attention mechanisms for the given task.

Bare soil will cause soil erosion and contribute to air pollution through the generation of dust, making the timely and effective monitoring of bare soil an urgent requirement for environmental management. Although there have been some researches on bare soil extraction using high-resolution remote sensing images, great challenges still need to be solved, such as complex background interference and small-scale problems. In this regard, the Hybrid Attention Network (HA-Net) is proposed for automatic extraction of bare soil from high-resolution remote sensing images, which includes the encoder and the decoder. In the encoder, HA-Net initially utilizes BoTNet for primary feature extraction, producing four-level features. The extracted highest-level features are then input into the constructed Spatial Information Perception Module (SIPM) and the Channel Information Enhancement Module (CIEM) to emphasize the spatial and channel dimensions of bare soil information adequately. To improve the detection rate of small-scale bare soil areas, during the decoding stage, the Semantic Restructuring-based Upsampling Module (SRUM) is proposed, which utilizes the semantic information from input features and compensate for the loss of detailed information during downsampling in the encoder. An experiment is performed based on high-resolution remote sensing images from the China–Brazil Resources Satellite 04A. The results show that HA-Net obviously outperforms several excellent semantic segmentation networks in bare soil extraction. The average precision and IoU of HA-Net in two scenes can reach 90.9% and 80.9%, respectively, which demonstrates the excellent performance of HA-Net. It embodies the powerful ability of HA-Net for suppressing the interference from complex backgrounds and solving multiscale issues. Furthermore, it may also be used to perform excellent segmentation tasks for other targets from remote sensing images.

Water detection from SAR imagery has significant values, such as the flood monitoring and environmental protection. Nowadays, significant progress has been achieved in water detection using deep neural network (DNN) methods, but the blackbox behavior incurs many doubts in the performance of deep learning techniques, which undermines its trustworthiness in water detection from SAR imagery. By integrating SAR domain knowledge, DNN and eXplainable Artificial Intelligence (XAI), an explainable DNN framework for surface water detection is proposed for the first time. This framework includes three parts: the water extraction network containing four backbone networks, the Local and Global Mixed Attribution (LGMA) module for performance evaluation of backbone network, and the Semantic Specific-class Activation Mapping (SSAM) module, which performs geo-visualization for the output layers of high-level features. In the experiment, SAR images from different resolutions and frequency-bands are utilized, which are from millimeter-wave and Sentinel-1 systems. The attribution maps and heatmaps of four backbone networks are assessed towards the final water extraction results. The experiment indicates that the proposed framework can glass-box the decision-making process of DNN in water detection and offer corresponding attribution analytics for given input SAR imagery. This work encourages other scholars to conduct extensive research on the explanation of DNN in SAR domain, gradually establish the trustworthiness of DNN, and promote the development of DNN in SAR images analytics.

PurposeTo evaluate the overall health status and health-related abilities and problems of elderly patients with diabetes and multimorbidity compared with those with diabetes only. Additionally, we aimed to identify different subgroups of elderly, multimorbid patients with diabetes.MethodsThis cross-sectional study included 538 elderly patients with diabetes. The participants completed a series of questionnaires on self-rated health (SRH), diabetes self-management, self-efficacy, health literacy, depression, and diabetes distress. Differences in health-related abilities and problems were compared between elderly patients with diabetes and multimorbidity and those with diabetes only, with adjustments for covariates using propensity score matching. A cluster analysis was also performed to identify the overall health status subgroups of elderly, multimorbid patients with diabetes. Additionally, we conducted a multinomial logistic regression analysis to examine the predictors of health-related abilities and problem-cluster group membership.ResultsElderly patients with diabetes and multimorbidity experienced more health-related abilities and problems than those with diabetes only, particularly within the domains of depression (p

In recent years, multibaseline phase unwrapping (MBPU) has been widely studied, but it is still suffering from the problem of low noise robustness. In view of this, two cluster correction methods for the cluster analysis-based MBPU algorithms are presented to improve the accuracy and effectiveness of the algorithm. In the first method, a suitable box centered on each pixel is selected and the cluster with the highest frequency in the box is taken as the corrected cluster to which the central pixel belongs; in the second method, all pixels are divided into core pixels and noncore pixels according to the predefined density, and then the cluster of the noncore pixels is corrected to the one that appears most frequently in the selected box centered on the noncore pixels. Experiments on both simulated and real datasets confirm the effectiveness of the proposed cluster correction methods.

This study introduces an interferometric synthetic aperture radar (InSAR) altimeter for sea-level elevation measurement. To evaluate the system capability and demonstrate that the InSAR altimeter can meet the application requirements of sea-surface elevation measurement, the main factors affecting relative elevation accuracy are analyzed. The mathematical expressions for each elevation measurement parameter and sea-surface elevation are derived along with the error transfer coefficients between each elevation measurement parameter and relative sea-surface elevation accuracy. Using the single-control variable method, the sensitivity of each elevation measurement parameter to relative elevation accuracy is obtained. By introducing the parameter values into the derivation formula of theoretical analysis, theoretical results are obtained. The theoretical results are compared with the simulation results, which helps to verify the validity of the error analysis. Results reveal that the interference phase error is the main factor affecting the accuracy of relative ocean height measurements. This work could significantly guide the subsequent design of the InSAR altimeter system.

Abstract Background Schisandra chinensis (Turcz.) Baill bee pollen extract (SCBPE) is often used as a functional food in China due to its good antioxidant property. However, its chemical compositions and effects on H9c2 cardiomyocytes against H2O2-induced cell injury still lacks of reports thus far. This study aimed to characterize the main components of SCBPE and investigate its protective effects against H2O2-induced H9c2 cardiomyocyte injury. Methods The main components of SCBPE were analyzed via ultraperformance liquid chromatography–quadrupole time-of-flight tandem mass spectrometry (UPLC–QTOF MS/MS). The three main nucleosides in SCBPE were quantitatively analyzed via ultraperformance liquid chromatography–diode array detection. Furthermore, the potential mechanism by which SCBPE exerts protective effects against H2O2-induced H9c2 cardiomyocyte injury was explored for the first time via cell survival rate measurements; cell morphological observation; myocardial superoxide dismutase (SOD) activity and malondialdehyde (MDA) and glutathione (GSH) level determination; flow cytometry; and quantitative polymerase chain reaction. Results Two carbohydrates, three nucleosides, and nine quinic acid nitrogen-containing derivatives in SCBPE were identified or tentatively characterized via UPLC–QTOF MS/MS. The nine quinic acid nitrogen-containing derivatives were first reported in bee pollen. The contents of uridine, guanosine, and adenosine were 2.4945 ± 0.0185, 0.1896 ± 0.0049, and 1.8418 ± 0.0157 μg/mg, respectively. Results of in vitro experiments showed that cell survival rate, myocardial SOD activity, and GSH level significantly increased and myocardial MDA level significantly decreased in SCBPE groups compared with those in H2O2 group. Cell morphology in SCBPE groups also markedly improved compared with that in H2O2 group. Results indicated that SCBPE protected H9c2 cardiomyocytes from H2O2-induced apoptosis by downregulating the mRNA expressions of Bax, cytochrome C, and caspase-3 and upregulating the Bcl-2 mRNA expression. Conclusions This study is the first to report that SCBPE could protect against oxidative stress injury and apoptosis in H2O2-injured H9c2 cells. Results indicated that the nucleosides and quinic acid nitrogen-containing derivatives could be the main substances that exert protective effects against H2O2-induced H9c2 cardiomyocyte injury.

The infrastructures built on soft clay in the Dongting Lake area are more prone to settlement and instability due to its significant rheological properties. It is of great importance to conduct a long-term surface deformation monitoring over this area. The most commonly used Synthetic Aperture Radar Interferometry (InSAR) deformation models are based on combinations of one or several pure empirical mathematical functions without considering the physical and mechanical characteristics of the observed objects. In this work, we propose an improved deformation model based on the functional relationship between strain and time in the Maxwell rheological model. The rheological parameters of elastic modulus and viscosity are introduced into a traditional empirical seasonal model. The improved model is applied for the investigation of the spatial-temporal surface evolution over the Dongting Lake area with the Small Baseline Subset InSAR (SBAS-InSAR) technology and TerraSAR-X satellite imagery. With the proposed model, the rheological parameters and the time series deformation are estimated, with the maximum accumulated subsidence estimated as 38 mm. Through the analysis of the generated results, we find that the lower the viscosity and elastic modulus are, the higher the deformation is. Temporally, the overall deformation follows a generally subsiding trend with a seasonal recovery of 5 mm from October 2012 to November 2012 and 12 mm from January 2013 to February 2013. To compensate for the deficiency of the unavailability of external geodetic measurements over this area, three different accuracy indexes (residual phase, temporal coherence, and high-pass deformation) are utilized to evaluate the modeling accuracy. The results of the improved model are also compared to three traditional models (seasonal model, cubic polynomial model, and linear model). The comparison shows that the improved model is highly recommended in this area because of its better accuracy.

The automatic extraction of airport runway areas from high-resolution Synthetic Aperture Radar (SAR) images is of great research significance in the military and civilian fields. However, it is still challenging to distinguish the airport from surrounding objects in SAR images. In this article, a new framework is proposed to extract airport runway areas (runways, taxiways, packing lots, and aircrafts) in a fast and automatic manner. The framework is based on the Geospatial Contextual Attention Mechanism (GCAM) for geospatial feature learning and classification, which is employed together with the down-sampling and coordinate mapping modules. To evaluate the performance of the proposed framework, three large-scale Gaofen-3 SAR images with 1m resolution are utilized in the experiment. According to the results, Mean Pixels Accuracy (MPA) and Mean Intersection Over Union (MIOU) of the GCAM are 0.9850 and 0.9536, respectively, which outperform RefineNet, DeepLabV3+, and MDDA. The training time of GCAM for the dataset is 2.25h, and the average testing time for the five SAR images is only 18.15s. Therefore, GCAM can offer rapid and automatic airport detection from high-resolution SAR images with high accuracy, which can further be employed to mark the airport to greatly improve the detection accuracy of the aircrafts.

Bee pollen possesses an anti-cardiomyocyte injury effect by reducing oxidative stress levels and inhibiting inflammatory response and apoptosis, but the possible effect mechanism has rarely been reported. This paper explores the effect of the extract of lotus bee pollen (LBPE) on cardiomyocyte hypertrophy (CH) and its mechanism. The main components of LBPE were identified via UPLC-QTOF MS. An isoproterenol-induced rat H9c2 CH model was subsequently used to evaluate the protection of LBPE on cells. LBPE (100, 250 and 500 μg∙mL−1) reduced the surface area, total protein content and MDA content, and increased SOD activity and GSH content in CH model in a dose-dependent manner. Meanwhile, quantitative real-time PCR trials confirmed that LBPE reduced the gene expression levels of CH markers, pro-inflammatory cytokines and pro-apoptosis factors, and increased the Bcl-2 mRNA expression and Bcl-2/Bax ratio in a dose-dependent manner. Furthermore, target fishing, bioinformatics analysis and molecular docking suggested JAK2 could be a pivotal target protein for the main active ingredients in the LBPE against CH. Ultimately, Western blot (WB) trials confirmed that LBPE can dose-dependently inhibit the phosphorylation of JAK2 and STAT3. The results show that LBPE can protect against ISO-induced CH, possibly via targeting the JAK2/STAT3 pathway, also suggesting that LBPE may be a promising candidate against CH.

Automatic river bridge detection is a typical and valuable application for SAR image analysis. However, the background information of SAR image is complex, and there are many specious targets with similar features, such as road, ponds and ridges, which usually cause false alarms. And current river bridge detection methods fail to handle these interference efficiently. Therefore, this paper applies deep learning to SAR and proposes a new river bridge detection algorithm, which is named as Single Short Detection-Adaptively Effective Feature Fusion (SSD-AEFF). It can effectively reduce the interference of noisy information, and achieve fast and high-precision detection of river bridges in complex SAR imagery. SSD-AEFF is based on SSD, and AEFF module is innovated to enhance the multi-scale feature maps together with effective Squeeze-Excitation (eSE) module to further fuse effective features and decrease the interference of background information. Additionally, Non-Maximum Suppression (NMS) is used to screen out redundant candidate boxes to produce the final detection result. Moreover, Gradient Harmonizing Mechanism (GHM) loss function is introduced to solve the problem of sample imbalance in the training process. Experimental results on TerraSAR data compared with existing baseline models demonstrate the superiority of the proposed SSD-AEFF algorithm.

Multibaseline (MB) phase unwrapping (PU) is a key processing technique in MB interferometric synthetic aperture radar (InSAR). As one of the most popular methods, the cluster analysis (CA)-based MBPU method often suffers from the problem of low noise robustness. Therefore, the block-matching and 3D filtering (BM3D) algorithm, one of the most effective filtering methods for image denoising, is applied to improve the performance of the method. Five different filtering strategies for applying BM3D are proposed in the paper: interferogram filtering (IFF), intercept filtering (ICF), cluster number filtering (CNF), unwrapped phase filtering (UPF), and simultaneous filtering (STF). In particular, while keeping the general structure of BM3D, four different similarity measures are defined for interferograms, intercepts, clusters, and unwrapped phases to accommodate the special characteristics of different filtering objects. Experiments on synthesized and real InSAR datasets prove their feasibility and effectiveness, and the experiment results show that (1) the PU accuracy and robustness of the CA-based MBPU method can be greatly improved by adding BM3D denoising; (2) simultaneous filtering of interferograms, intercepts, cluster numbers, and unwrapped phases works best, but with the worst time complexity; (3) when filtering is performed for only one object of the CA-based MBPU method, the filtering effect of CNF and UPF is better than that of IFF and ICF; and (4), considering the three indicators of PUSR, NRSE, and time consumption, CNF and UPF should be the best choices.

Deformation prediction for a salt solution mining area is essential to mining environmental protection. The combination of Synthetic Aperture Radar Interferometry (InSAR) technique with Probability Integral Method (PIM) has proven to be powerful in predicting mining-induced subsidence. However, traditional mathematical empirical models (such as linear model or linear model combined with periodical function) are mostly used in InSAR approaches, ignoring the underground mining mechanisms, which may limit the accuracy of the retrieved deformations. Inaccurate InSAR deformations will transmit an unavoidable error to the estimated PIM parameters and the forward predicted subsidence, which may induce more significant errors. Besides, theoretical contradictory and non-consistency between InSAR deformation model and future prediction model is another limitation. This paper introduces the Coordinate-Time (CT) function into InSAR deformation modeling. A novel time-series InSAR model (namely, CT-PIM) is proposed as a substitute for traditional InSAR mathematical empirical models and directly applied for future dynamic prediction. The unknown CT-PIM parameters can be estimated directly via InSAR phase observations, which can avoid the error propagation from the InSAR-generated deformations. The new approach has been tested by both simulated and real data experiments over a salt mine in China. The root mean square error (RMSE) is determined as ±10.9 mm, with an improvement of 37.2% compared to traditional static PIM prediction method. The new approach provides a more robust tool for the forecasting of mining-induced hazards in salt solution mining areas, as well as a reference for ensuring the environment protection and safety management.

Although deep learning has achieved great success in aircraft detection from SAR imagery, its blackbox behavior has been criticized for low comprehensibility and interpretability. Such challenges have impeded the trustworthiness and wide application of deep learning techniques in SAR image analytics. In this paper, we propose an innovative eXplainable Artificial Intelligence (XAI) framework to glassbox deep neural networks (DNN) by using aircraft detection as a case study. This framework is composed of three parts: hybrid global attribution mapping (HGAM) for backbone network selection, path aggregation network (PANet), and class-specific confidence scores mapping (CCSM) for visualization of the detector. HGAM integrates the local and global XAI techniques to evaluate the effectiveness of DNN feature extraction; PANet provides advanced feature fusion to generate multi-scale prediction feature maps; while CCSM relies on visualization methods to examine the detection performance with given DNN and input SAR images. This framework can select the optimal backbone DNN for aircraft detection and map the detection performance for better understanding of the DNN. We verify its effectiveness with experiments using Gaofen-3 imagery. Our XAI framework offers an explainable approach to design, develop, and deploy DNN for SAR image analytics.

Gaze-following is an effective way for intention understanding in human–robot interaction, which aims to follow the gaze of humans to estimate what object is being observed. Most of the existing methods require people and objects to appear in the same image. Due to the limitation in the view of the camera, these methods are not applicable in practice. To address this problem, we propose a method of gaze following that utilizes a geometric map for better estimation. With the help of the map, this method is competitive for cross-frame estimation. On the basis of this method, we propose a novel gaze-based image caption system, which has been studied for the first time. Our experiments demonstrate that the system follows the gaze and describes objects accurately. We believe that this system is competent for autistic children’s rehabilitation training, pension service robots, and other applications.

In aircraft detection from synthetic aperture radar (SAR) images, there are several major challenges: the shattered features of the aircraft, the size heterogeneity and the interference of a complex background. To address these problems, an Efficient Bidirectional Path Aggregation Attention Network (EBPA2N) is proposed. In EBPA2N, YOLOv5s is used as the base network and then the Involution Enhanced Path Aggregation (IEPA) module and Effective Residual Shuffle Attention (ERSA) module are proposed and systematically integrated to improve the detection accuracy of the aircraft. The IEPA module aims to effectively extract advanced semantic and spatial information to better capture multi-scale scattering features of aircraft. Then, the lightweight ERSA module further enhances the extracted features to overcome the interference of complex background and speckle noise, so as to reduce false alarms. To verify the effectiveness of the proposed network, Gaofen-3 airports SAR data with 1 m resolution are utilized in the experiment. The detection rate and false alarm rate of our EBPA2N algorithm are 93.05% and 4.49%, respectively, which is superior to the latest networks of EfficientDet-D0 and YOLOv5s, and it also has an advantage of detection speed.

The automatic detection of aircrafts from SAR images is widely applied in both military and civil fields, but there are still considerable challenges. To address the high variety of aircraft sizes and complex background information in SAR images, a new fast detection framework based on convolution neural networks is proposed, which achieves automatic and rapid detection of aircraft with high accuracy. First, the airport runway areas are detected to generate the airport runway mask and rectangular contour of the whole airport are generated. Then, a new deep neural network proposed in this paper, named Efficient Weighted Feature Fusion and Attention Network (EWFAN), is used to detect aircrafts. EWFAN integrates the weighted feature fusion module, the spatial attention mechanism, and the CIF loss function. EWFAN can effectively reduce the interference of negative samples and enhance feature extraction, thereby significantly improving the detection accuracy. Finally, the airport runway mask is applied to the detected results to reduce false alarms and produce the final aircraft detection results. To evaluate the performance of the proposed framework, large-scale Gaofen-3 SAR images with 1 m resolution are utilized in the experiment. The detection rate and false alarm rate of our EWFAN algorithm are 95.4% and 3.3%, respectively, which outperforms Efficientdet and YOLOv4. In addition, the average test time with the proposed framework is only 15.40 s, indicating satisfying efficiency of automatic aircraft detection.

Water detection from Synthetic Aperture Radar (SAR) images has been widely utilized in various applications. However, it remains an open challenge due to the high similarity between water and shadow in SAR images. To address this challenge, a new end-to-end framework based on deep learning has been proposed to automatically classify water and shadow areas in SAR images. This end-to-end framework is mainly composed of three parts, namely, Multi-scale Spatial Feature (MSF) extraction, Multi-Level Selective Attention Network (MLSAN) and the Improvement Strategy (IS). Firstly, the dataset is input to MSF for multi-scale low-level feature extraction via three different methods. Then, these low-level features are fed into the MLSAN network, which contains the Encoder and Decoder. The Encoder aims to generate different levels of features using residual network of 101 layers. The Decoder extracts geospatial contextual information and fuses the multi-level features to generate high-level features that are further optimized by the IS. Finally, the classification is implemented with the Softmax function. We name the proposed framework as MSF-MLSAN, which is trained and tested using millimeter wave SAR datasets. The classification accuracy reaches 0.8382 and 0.9278 for water and shadow, respectively; while the overall Intersection over Union (IoU) is 0.9076. MSF-MLSAN demonstrates the success of integrating SAR domain knowledge and state-of-the-art deep learning techniques.

The detection of airports from Synthetic Aperture Radar (SAR) images is of great significance in various research fields. However, it is challenging to distinguish the airport from surrounding objects in SAR images. In this paper, a new framework, multi-level and densely dual attention (MDDA) network is proposed to extract airport runway areas (runways, taxiways, and parking lots) in SAR images to achieve automatic airport detection. The framework consists of three parts: down-sampling of original SAR images, MDDA network for feature extraction and classification, and up-sampling of airports extraction results. First, down-sampling is employed to obtain a medium-resolution SAR image from the high-resolution SAR images to ensure the samples (500 × 500) can contain adequate information about airports. The dataset is then input to the MDDA network, which contains an encoder and a decoder. The encoder uses ResNet_101 to extract four-level features with different resolutions, and the decoder performs fusion and further feature extraction on these features. The decoder integrates the chained residual pooling network (CRP_Net) and the dual attention fusion and extraction (DAFE) module. The CRP_Net module mainly uses chained residual pooling and multi-feature fusion to extract advanced semantic features. In the DAFE module, position attention module (PAM) and channel attention mechanism (CAM) are combined with weighted filtering. The entire decoding network is constructed in a densely connected manner to enhance the gradient transmission among features and take full advantage of them. Finally, the airport results extracted by the decoding network were up-sampled by bilinear interpolation to accomplish airport extraction from high-resolution SAR images. To verify the proposed framework, experiments were performed using Gaofen-3 SAR images with 1 m resolution, and three different airports were selected for accuracy evaluation. The results showed that the mean pixels accuracy (MPA) and mean intersection over union (MIoU) of the MDDA network was 0.98 and 0.97, respectively, which is much higher than RefineNet and DeepLabV3. Therefore, MDDA can achieve automatic airport extraction from high-resolution SAR images with satisfying accuracy.

Bridge detection from Synthetic Aperture Radar (SAR) images has very important strategic significance and practical value, but there are still many challenges in end-to-end bridge detection. In this paper, a new deep learning-based network is proposed to identify bridges from SAR images, namely, multi-resolution attention and balance network (MABN). It mainly includes three parts, the attention and balanced feature pyramid (ABFP) network, the region proposal network (RPN), and the classification and regression. First, the ABFP network extracts various features from SAR images, which integrates the ResNeXt backbone network, balanced feature pyramid, and the attention mechanism. Second, extracted features are used by RPN to generate candidate boxes of different resolutions and fused. Furthermore, the candidate boxes are combined with the features extracted by the ABFP network through the region of interest (ROI) pooling strategy. Finally, the detection results of the bridges are produced by the classification and regression module. In addition, intersection over union (IOU) balanced sampling and balanced L1 loss functions are introduced for optimal training of the classification and regression network. In the experiment, TerraSAR data with 3-m resolution and Gaofen-3 data with 1-m resolution are used, and the results are compared with faster R-CNN and SSD. The proposed network has achieved the highest detection precision (P) and average precision (AP) among the three networks, as 0.877 and 0.896, respectively, with the recall rate (RR) as 0.917. Compared with the other two networks, the false alarm targets and missed targets of the proposed network in this paper are greatly reduced, so the precision is greatly improved.

The water and shadow areas in SAR images contain rich information for various applications, which cannot be extracted automatically and precisely at present. To handle this problem, a new framework called Multi-Resolution Dense Encoder and Decoder (MRDED) network is proposed, which integrates Convolutional Neural Network (CNN), Residual Network (ResNet), Dense Convolutional Network (DenseNet), Global Convolutional Network (GCN), and Convolutional Long Short-Term Memory (ConvLSTM). MRDED contains three parts: the Gray Level Gradient Co-occurrence Matrix (GLGCM), the Encoder network, and the Decoder network. GLGCM is used to extract low-level features, which are further processed by the Encoder. The Encoder network employs ResNet to extract features at different resolutions. There are two components of the Decoder network, namely, the Multi-level Features Extraction and Fusion (MFEF) and Score maps Fusion (SF). We implement two versions of MFEF, named MFEF1 and MFEF2, which generate separate score maps. The difference between them lies in that the Chained Residual Pooling (CRP) module is utilized in MFEF2, while ConvLSTM is adopted in MFEF1 to form the Improved Chained Residual Pooling (ICRP) module as the replacement. The two separate score maps generated by MFEF1 and MFEF2 are fused with different weights to produce the fused score map, which is further handled by the Softmax function to generate the final extraction results for water and shadow areas. To evaluate the proposed framework, MRDED is trained and tested with large SAR images. To further assess the classification performance, a total of eight different classification frameworks are compared with our proposed framework. MRDED outperformed by reaching 80.12% in Pixel Accuracy (PA) and 73.88% in Intersection of Union (IoU) for water, 88% in PA and 77.11% in IoU for shadow, and 95.16% in PA and 90.49% in IoU for background classification, respectively.

Building deformation models consistent with reality is a crucial step for time-series deformation monitoring. Most deformation models are empirical mathematical models, lacking consideration of the physical mechanisms of observed objects. In this study, we propose an improved time-series deformation model considering rheological parameters (viscosity and elasticity) based on the Kelvin model. The functional relationships between the rheological parameters and deformation along the Synthetic Aperture Radar ( SAR) line of sight are constructed, and a method for rheological parameter estimation is provided. To assess the feasibility and accuracy of the presented model, both simulated and real deformation data over a stretch of the Lungui highway (built on soft clay subgrade in Guangdong province, China) are investigated with TerraSAR-X satellite imagery. With the proposed deformation model, the unknown rheological parameters over all the high coherence points are obtained and the deformation time-series are generated. The high-pass (HP) deformation component and external leveling ground measurements are utilized to assess the modeling accuracy. The results show that the root mean square of the residual deformation is ±1.6 mm, whereas that of the ground leveling measurements is ±5.0 mm, indicating an improvement in the proposed model by 53%, and 34% compared to the pure linear velocity model. The results indicate the reliability of the presented model for the application of deformation monitoring of soft clay highways. The estimated rheological parameters can be provided as a reference index for the interpretation of long-term highway deformation and the stability control of subgrade construction engineering.

Highways built on soft clay subgrade are more prone to subsidence due to the geotechnical characteristics of soft clay. Monitoring ground movements in this area is significant for understanding the deformation dynamics and reducing maintenance cost as well. In this paper, small baseline subset synthetic aperture radar interferometry (SBAS-InSAR) technique is exploited to obtain and investigate the time series ground surface deformation after the construction of a road embankment over soft clay settlement. Considering the important effect of temporal deformation models on the final accuracy of estimated deformation, both the linear velocity model and seasonal deformation model are utilized to conduct the comparative investigation of deformation time series. Two highways in Fuoshan, China—G1501 Guangzhou Belt Highway and Lungui Highway—were selected as the test area. Thirteen TerraSAR-X images acquired from October 2014 to November 2015 were analyzed. Comparative study based on two groups of analyses generated from the two models for both highways were conducted. Consequently, several feature points distributed near the two highways were analyzed in detail to understand the temporal evolution of the settlement. In order to evaluate the reliability of our measurements, the residual phase was analyzed to assess the modelling accuracy of the two models. In addition, leveling data were also used to validate the experimental results. Our measurements suggest that the seasonal model is more suitable for the test highways, with an accuracy of ±3 mm with respect to the leveling results.

Synthetic Aperture Radar (SAR) scene classification is challenging but widely applied, in which deep learning can play a pivotal role because of its hierarchical feature learning ability. In the paper, we propose a new scene classification framework, named Feature Recalibration Network with Multi-scale Spatial Features (FRN-MSF), to achieve high accuracy in SAR-based scene classification. First, a Multi-Scale Omnidirectional Gaussian Derivative Filter (MSOGDF) is constructed. Then, Multi-scale Spatial Features (MSF) of SAR scenes are generated by weighting MSOGDF, a Gray Level Gradient Co-occurrence Matrix (GLGCM) and Gabor transformation. These features were processed by the Feature Recalibration Network (FRN) to learn high-level features. In the network, the Depthwise Separable Convolution (DSC), Squeeze-and-Excitation (SE) Block and Convolution Neural Network (CNN) are integrated. Finally, these learned features will be classified by the Softmax function. Eleven types of SAR scenes obtained from four systems combining different bands and resolutions were trained and tested, and a mean accuracy of 98.18% was obtained. To validate the generality of FRN-MSF, five types of SAR scenes sampled from two additional large-scale Gaofen-3 and TerraSAR-X images were evaluated for classification. The mean accuracy of the five types reached 94.56%; while the mean accuracy for the same five types of the former tested 11 types of scene was 96%. The high accuracy indicates that the FRN-MSF is promising for SAR scene classification without losing generality.

Oxidative stress plays an important role in the pathogenesis of myocardial infarction (MI). Schisandra chinensis bee pollen extract (SCBPE) possesses powerful antioxidant capacity. This study aimed to further explore the antioxidative and cardioprotective effects of SCBPE on acute MI induced by isoprenaline (ISO) in rats. The rats were intragastrically administrated with SCBPE (600, 1200, or 1800 mg/kg/day) and Compound Danshen dropping pills (270 mg/kg/day) for 30 days, then subcutaneously injected with ISO (65 mg/kg/day) on the 29th and 30th day. Compared with the model group, pretreatment with middle and high doses of SCBPE significantly reduced serum aspartate transaminase, lactate dehydrogenase, and creatine kinase activities and increased myocardial superoxide dismutase, glutathione peroxidase, and catalase activities. The histopathologic aspects showed that pathological heart change was found in the model group and reduced to varying degrees in the SCBPE groups. Moreover, the protein expression of nuclear factor-erythroid 2-related factor 2 (Nrf-2), heme oxygenase-1 (HO-1), and Bcl2 in the heart increased in the SCBPE groups, while that of Bax decreased compared to the model group. Besides this, uridine was isolated from S. chinensis bee pollen for the first time. This study could provide a scientific basis for using Schisandra chinensis bee pollen as a functional food for the prevention of MI.

Monitoring surface movement near highways over soft clay subgrades is fundamental for understanding the dynamics of the settlement process and preventing hazards. Earlier studies have demonstrated the accuracy and cost-effectiveness of using time series radar interferometry (InSAR) technique to measure the ground deformation. However, the accuracy of the advanced differential InSAR techniques, including short baseline subset (SBAS) InSAR, is limited by the temporal deformation models used. In this study, a comparison of four widely used time series deformation models in InSAR, namely Multi Velocity Model (MVM), Permanent Velocity Model (PVM), Seasonal Model (SM) and Cubic Polynomial Model (CPM), was conducted to measure the long-term ground deformation after the construction of road embankment over soft clay subgrade. SBAS-InSAR technique with TerraSAR-X satellite imagery were conducted to generate the time series deformation data over the studied highway. In the experiments, three accuracy indices were applied to show the residual phase, mean temporal coherence and the RMS of high-pass deformation, respectively. In addition, the derived time series deformation maps of the highway based on the four selected models and 17 TerraSAR-X images acquired from June 2014 to November 2015 were compared. The leveling data was also used to validate the experimental results. Our results suggested the Seasonal Model is the most suitable model for the selected study site. Consequently, we analyzed two bridges in detail and three single points distributed near the highway. Compared with the ground leveling deformation measurements and results of other models, SM showed better consistency, with the accuracy of deformation to be ±7 mm.