Search

Although existing anchor-based oriented object detection methods have achieved remarkable results, they require manual preset boxes, which introduce additional hyper-parameters and calculations. These methods often use more complex architectures for better performance, which makes them difficult to deploy on computationally constrained embedded platforms, such as satellites and unmanned aerial vehicles. We aim to design a high-performance algorithm that is simple, fast, and easy to deploy for aerial image detection. In this article, we propose a one-stage anchor-free rotated object detector, FCOSR, that can be deployed on most platforms and uses our well-defined label assignment strategy for the features of the aerial image objects. We use the ellipse center sampling method to define a suitable sampling region for an oriented bounding box (OBB). The fuzzy sample assignment strategy provides reasonable labels for overlapping objects. To solve the problem of insufficient sampling, we designed a multi-level sampling module. These strategies allocate more appropriate labels to training samples. Our algorithm achieves an mean average precision (mAP) of 79.25, 75.41, and 90.13 on the DOTA-v1.0, DOTA-v1.5, and HRSC2016 datasets, respectively. FCOSR demonstrates a performance superior to that of other methods in single-scale evaluation, where the small model achieves an mAP of 74.05 at a speed of 23.7 FPS on an RTX 2080-Ti GPU. When we convert the lightweight FCOSR model to the TensorRT format, it achieves an mAP of 73.93 on DOTA-v1.0 at a speed of 17.76 FPS on a Jetson AGX Xavier device with a single scale. [ABSTRACT FROM AUTHOR]

With the development of remote sensing earth observation technology, object tracking has gained attention for its broad application prospects in computer vision. However, object tracking is challenging owing to the background clutter, occlusion, and scale variation that often appear in remote sensing videos. Many existing trackers cannot accurately track the object for remote sensing videos with complex backgrounds. Several tracking methods can handle just one situation, such as occlusion. In this article, we propose a Siamese multi-scale adaptive search (SiamMAS) network framework to achieve object tracking for remote sensing videos. First, a multi-scale cross correlation is presented to obtain a more discriminative model and comprehensive feature representation, improving the performance of the model to handle complex backgrounds in remote sensing videos. Second, an adaptive search module is employed that augments the Kalman filter with a partition search strategy for object motion estimation. The Kalman filter is adopted to re-detect the object when the network cannot track the object in the current frame. Moreover, the partition search strategy can help the Kalman filter accomplish a more accurate region-proposal selection. Finally, extensive experiments on remote sensing videos taken from Jilin-1 commercial remote sensing satellites show that the proposed tracking algorithm achieves strong tracking performance with 0.913 precision while running at 37.528 frames per second (FPS), demonstrating its effectiveness and efficiency. [ABSTRACT FROM AUTHOR]

Diabetes mellitus (DM) is a severe chronic diseases with a global prevalence of 9%, leading to poor health and high health care costs, and is a direct cause of millions of deaths each year. The rising epidemic of diabetes and its complications, such as retinal and peripheral nerve disease, is a huge burden globally. A better understanding of the molecular pathways involved in the development and progression of diabetes and its complications can facilitate individualized prevention and treatment. High diabetes mellitus incidence rate is caused mainly by lack of non-invasive and reliable methods for early diagnosis, such as plasma biomarkers. The incidence of diabetes and its complications in the world still grows so it is crucial to develop a new, faster, high specificity and more sensitive diagnostic technologies. With the advancement of analytical techniques, metabolomics can identify and quantify multiple biomarkers simultaneously in a high-throughput manner, and effective biomarkers can greatly improve the efficiency of diabetes and its complications. By providing information on potential metabolic pathways, metabolomics can further define the mechanisms underlying the progression of diabetes and its complications, help identify potential therapeutic targets, and improve the prevention and management of T2D and its complications. The application of amino acid metabolomics in epidemiological studies has identified new biomarkers of diabetes mellitus (DM) and its complications, such as branched-chain amino acids, phenylalanine and arginine metabolites. This study focused on the analysis of metabolic amino acid profiling as a method for identifying biomarkers for the detection and screening of diabetes and its complications. The results presented are all from recent studies, and in all cases analyzed, there were significant changes in the amino acid profile of patients in the experimental group compared to the control group. This study demonstrates the potential of amino acid profiles as a detection method for diabetes and its complications. [ABSTRACT FROM AUTHOR]

Objective: The repair of great toe donor site defect after wrap‐around flap transfer is still controversial. The bilobed superficial circumflex iliac artery perforator (SCIP) flap can improve the aesthetics of the great toe while maintaining its function. Thus, this study aimed to report our experience in the reconstruction of big toe donor site defects with the bilobed SCIP flap and describe the clinical outcomes. Methods: This study was a retrospective trial. From May 2017 to May 2020, 13 patients with the great toe donor site defect after wrap‐around flap transfer were included in this study. The average age of the patients was 44 years (range, 23–60 years). All patients received free bilobed SCIP flaps to reconstruct the donor site defect of the great toe. Relevant clinical features were recorded preoperatively. The thickness and design of the SCIP flap and the harvesting layer of the flap were measured during the operation. The survival rate of flaps and skin grafts and the incidence of infection were recorded after operation. At follow‐up, donor site complications and postoperative outcomes were evaluated. Results: In all cases, the SCIP flap covering the donor site of the great toe survived. All patients were followed up for 24–40 months (mean, 30.5 months). The average thickness of the SCIP flap was 0.38cm. All SCIP flaps were harvested from the superficial fascial layer except for three obese patients. The thin SCIP flap had a bilobed design with no further defatting procedures. Postoperatively, the great toe‐nail flap donor site regained its original appearance without bloating or flap necrosis. There was a hidden linear scar in the groin donor site, which did not affect hip joint movement. All patients were satisfied with the aesthetics of the surgical site. Conclusion: The SCIP flap with bilobed design for repairing the donor defect of the great toe after wrap‐around flap transfer is a kind of surgical method with excellent contour, meeting the requirements of function and aesthetics. [ABSTRACT FROM AUTHOR]

Capturing the dependencies between joints is critical in skeleton-based action recognition. However, the existing methods cannot effectively capture the correlation of different joints between frames, which is very useful since different body parts (such as the arms and legs in "long jump") between adjacent frames move together. Focus on this issue, a novel spatio-temporal segments attention method is proposed. The skeleton sequence is divided into several segments, and several consecutive frames contained in each segment are encoded. And then an intra-segment self-attention module is proposed to capture the relationship of different joints in consecutive frames. In addition, an inter-segment action attention module is introduced to capture the relationship between segments to enhance the ability to distinguish similar actions. Compared with the state-of-the-art methods, our method achieves better performance on two large-scale datasets. [ABSTRACT FROM AUTHOR]

Traditional clustering algorithms such as K-means are highly sensitive to the initial centroid selection and perform poorly on non-convex dataset. To address these problems, a novel data adaptive Bregman clustering algorithm (DBGSA) is proposed in this paper, which combines the universal gravitational algorithm to bring similar points closer in the dataset. In addition, we introduce the Bregman divergence generalized power mean information loss minimization to identify cluster centers and build a hyperparameter identification optimization model, which effectively solves the problems of manual adjustment and uncertainty in the improved dataset. Extensive experiments are conducted on four simulated datasets and six real datasets. The results demonstrate that DBGSA significantly improves the accuracy of various clustering algorithms by an average of 63.8% compared to the contrasted algorithms, making it applicable to diverse datasets. [ABSTRACT FROM AUTHOR]

During the past decades, convolutional neural network (CNN)-based models have achieved notable success in remote sensing image classification due to their powerful feature representation ability. However, the lack of explainability during the decision-making process is a common criticism of these high-capacity networks. Local explanation methods that provide visual saliency maps have attracted increasing attention as a means to surmount the barrier of explainability. However, the vast majority of research is conducted on the last convolutional layer, where the salient regions are unintelligible for partial remote sensing images, especially scenes that contain plentiful small targets or are similar to the texture image. To address these issues, we propose a novel framework called Prob-POS, which consists of the class-activation map based on the probe network (Prob-CAM) and the weighted probability of occlusion (wPO) selection strategy. The proposed probe network is a simple but effective architecture to generate elaborate explanation maps and can be applied to any layer of CNNs. The wPO is a quantified metric to evaluate the explanation effectiveness of each layer for different categories to automatically pick out the optimal explanation layer. Variational weights are taken into account to highlight the high-scoring regions in the explanation map. Experimental results on two publicly available datasets and three prevalent networks demonstrate that Prob-POS improves the faithfulness and explainability of CNNs on remote sensing images. [ABSTRACT FROM AUTHOR]

Touzi decomposition provides more scattering information corresponding to parameters describing different targets. It is possible to explore the selection method of parameters, which plays an important role in polarimetric synthetic aperture radar (PolSAR) image classification. Therefore, this paper presents an innovative parameter ordering scheme based on the histogram of parameters obtained by Touzi decomposition. Then, a parameter selection scheme is proposed by analyzing the influence of the increase of parameters on the overall accuracy obtained by the Softmax classifier. Based on the Gamma distribution, the parameter selection method of eigenvalue decomposition is also given. As the data distribution of the selected parameters does not always satisfy the Gaussian distribution, it may not be sufficient to directly apply the autoencoder network. Thus, the loss function of the autoencoder network is modified by the construction of different data error terms according to a different data distribution form of the selected parameters, and then an improved autoencoder network is proposed. The process of feature extraction and classification can be regarded as a whole network to better accomplish the task of classification. An extensive set of experiments are done on four real PolSAR images. Compared with the classification results of all parameters, the gap in the overall accuracy of the parameters obtained by the proposed selection scheme is only about 1%. In terms of the classification overall accuracy, considering the distribution of parameters in the autoencoder network and the construction of the whole classification network can improve it by about 6–7% for four data sets. Code is available at https://github.com/Justin20220123/ISPRS2022. [ABSTRACT FROM AUTHOR]

Weakly supervised object localization (WSOL) is a challenging task that aims to localize objects in images using only image-level labels. Despite the widespread use of WSOL methods based on class activation mapping (CAM), such methods do not consider that the network may overly focus on local regions of the most interesting objects during localization, thus neglecting the overall information. To address this issue, we introduces an additional attention branch for convolutional neural networks (CNNs) that utilizes the attention mechanism of multi-layer perceptron (MLP) to enhance the network's learning of global information and supervise the feature learning of CNNs online through knowledge distillation, thereby improving the localization accuracy of WSOL. Specifically, we designs a new loss function using the generated features to combine with contrastive learning, effectively dividing the foreground and background of the image to provide more accurate pseudo-labels for subsequent classification and localization tasks. In the experiments, we tested our method on the CUB-200-2011 dataset and compared it with existing methods. The experimental results show that our method achieves good performance in WSOL tasks. [ABSTRACT FROM AUTHOR]

Joint-level and part-level information are crucial for modeling actions with different granularity. In addition, the relevant information on different joints between consecutive frames is very useful for skeleton-based action recognition. To effectively capture the action information, a new multi-grained clip focus network (MGCF-Net) is proposed. Firstly, the skeleton sequence is divided into multiple clips, each containing several consecutive frames. According to the structure of the human body, each clip is divided into several tuples. Then an intra-clip attention module is proposed to capture intra-clip action information. Specifically, multi-head self-attention is divided into two parts, obtaining relevant information at the joint and part levels, and integrating the information captured from these two parts to obtain multi-grained contextual features. In addition, an inter-clip focus module is used to capture the key information of several consecutive sub-actions, which will help to distinguish similar actions. On two large-scale benchmarks for skeleton-based action recognition, our method achieves the most advanced performance, and its effectiveness has been verified. • A clip encoding strategy is proposed to mine spatiotemporal joint information. • An intra-clip attention module is proposed to capture multi-grained information within each clip. • An inter-clip focus module is proposed to focus on key sub-actions. • The experiment demonstrates the performance and effectiveness of our method. [ABSTRACT FROM AUTHOR]

Convolutional neural network (CNN)-based research has been successfully applied in remote sensing image classification due to its powerful feature representation ability. However, these high-capacity networks bring heavy inference costs and are easily overparameterized, especially for the deep CNNs pretrained on natural image datasets. Network pruning is regarded as a prevalent approach for compressing networks, but most existing research ignores model interpretability while formulating pruning criterion. To address these issues, a network pruning method for remote sensing image classification based on interpretable CNNs is proposed. More specifically, an original interpretable CNN with a predefined pruning ratio is trained at first. The filters, namely channels in the high convolutional layer, are able to learn specific semantic meanings in proportion to the predefined pruning ratio. The filters without interpretability are supposed to be removed. As for other convolutional layers, a sensitivity function is designed to assess the risk of pruning channels for each layer, and furthermore, the pruning ratio for each layer is corrected adaptively. The pruning method based on the proposed sensitivity function is effective and requires little computational costs to search abandoned channels without damaging classification performance. To demonstrate the effectiveness, the proposed method is implemented on different scales of modern CNN models, including VGG-VD and AlexNet. The experimental results, obtained on the UC Merced dataset and NWPU-RESISC45 dataset, prove that our method significantly reduces the inference costs and improves the interpretability of networks. [ABSTRACT FROM AUTHOR]

Objective: This study has been compared the effectiveness of different surgical methods in the treatment of mitral regurgitation (MR) in adults by using the network meta‐analysis method, so as to provide reference for clinical selection of the best surgical scheme. Methods: The PubMed, EMBASE, the Cochrane Library, CNKI, and Chongqing VIP Information databases were comprehensively searched until December 2020. We collected retrospective comparative studies on surgical procedures including 3D endoscopic mitral valve surgery (3D‐MVS), robot‐assisted mitral valve surgery (R‐MVS); totally thoracoscopic mitral valve surgery (T‐MVS), small incision mitral valve surgery (M‐MVS), and traditional thoracotomy mitral valve surgery (C‐MVS). Stata16.0 and Addis1.16.8 software was used for network meta‐analysis using the Bayesian approach. Results: A total of 31 studies were included, 12,998 patients, involving five surgical methods. Network meta‐analysis showed that: in terms of complications (odds ratio [OR]: 0.65, 95% CI: 0.13–3.00, probability rank = 0.37) and MR (OR: 0.03, 95% CI: 0.0–8315, probability rank = 0.64), the 3D‐MVS group had the lowest event rate. In terms of blood transfusion rate (OR: 0.55, 95% CI: 0.16–1.84, probability rank = 0.45), T‐MVS had the lowest event rate. In addition, with the exception of operation time and chest drainage, the R‐MVS group has the best curative effect. Conclusion: This minimally invasive surgery has their own advantages and disadvantages. Overall, 3D‐MVS is most satisfactory, but more samples are needed. [ABSTRACT FROM AUTHOR]

This final number will be Graph HT ht . This final limit of this symbolic expression as symbolic variables approach the original values in matrix I A i will be Graph HT ht . Its determinant is denoted by Graph HT ht , also written Graph HT ht . Next, let us look at the following specific example: Graph HT ht where the I a i changes from 2 to -3, the I b i changes from 0 to 10. [Extracted from the article]

Land cover classification is an important application for polarimetric synthetic aperture radar (PolSAR) image interpretation. The classification performance of a promising parametric feature and classifier learning-based algorithm is limited when the amounts of pixels from different classes vary greatly. PolSAR data from minority classes is difficult to recognize correctly owing to a strong learning bias toward the majority classes, resulting in under-performing features for minority classes. To address this issue, a cost-sensitive latent space learning network based on the feature and classifier learning framework is proposed to reduce the learning bias for supporting the classification of imbalanced data in PolSAR images. First, a new cost-sensitive method is developed by adaptively computing the cost coefficient from predicted labels in the optimization process. Thus, the imbalanced distribution of PolSAR data can be obtained for both the labeled and unlabeled pixels rather than a predefined misclassifying matrix for labeled pixels. Second, latent space learning is used as an auxiliary task to assist the main task of classifier learning. By weighting the distance between the learned feature and the basis of the latent space with a different cost-sensitive coefficient, pixels in minority and majority classes are promoted to be more separable. Thus, the strong bias to majority classes is reduced from both the feature learning and classification process. Finally, the proposed method is studied through experiments on three different PolSAR images with several existing state-of-the-art methods. The experiments validate the effectiveness of the proposed method for balanced and imbalanced PolSAR land cover classification. [ABSTRACT FROM AUTHOR]

Effective features play an important role in synthetic aperture radar (SAR) image interpretation. However, since SAR images contain a variety of terrain types, it is not easy to extract effective features of different terrains from SAR images. Deep learning methods require a large amount of labeled data, but the difficulty of SAR image annotation limits the performance of deep learning models. SAR images have inevitable geometric distortion and coherence speckle noise, which makes it difficult to extract effective features from SAR images. If effective semantic context features cannot be learned for SAR images, the extracted features struggle to distinguish different terrain categories. Some existing terrain classification methods are very limited and can only be applied to some specified SAR images. To solve these problems, a jigsaw puzzle self-supervised learning (JPSSL) framework is proposed. The framework comprises a jigsaw puzzle pretext task and a terrain classification downstream task. In the pretext task, the information in the SAR image is learned by completing the SAR image jigsaw puzzle to extract effective features. The terrain classification downstream task is trained using only a small number of labeled data. Finally, fully connected conditional random field processing is performed to eliminate noise points and obtain a high-quality terrain classification result. Experimental results on three large-scene high-resolution SAR images confirm the effectiveness and generalization of our method. Compared with the supervised methods, the features learned in JPSSL are highly discriminative, and the JPSSL achieves good classification accuracy when using only a small amount of labeled data. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) image classification is a fundamental research direction in image interpretation. With the development of various intelligent technologies, deep learning techniques are gradually being applied to SAR image classification. In this study, a new SAR classification algorithm known as the multiscale convolutional neural network with an autoencoder regularization joint contextual attention network (MCAR-CAN) is proposed. The MCAR-CAN has two branches: the autoencoder regularization branch and the context attention branch. First, autoencoder regularization is used for the reconstruction of the input to regularize the classification in the autoencoder regularization branch. Multiscale input and an asymmetric structure of the autoencoder branch cause the network more to be focused on classification than on reconstruction. Second, the attention mechanism is used to produce an attention map in which each attention weight corresponds to a context correlation in attention branch. The robust features are obtained by the attention mechanism. Finally, the features obtained by the two branches are spliced for classification. In addition, a new training strategy and a postprocessing method are designed to further improve the classification accuracy. Experiments performed on the data from three SAR images demonstrated the effectiveness and robustness of the proposed algorithm. [ABSTRACT FROM AUTHOR]

In this paper, a lopsided modified Euler-extrapolated Hermitian and skew-Hermitian splitting (LME-HS) iteration method is introduced for solving the complex symmetric linear systems. Under a loose restriction on parameter θ, we demonstrate that LME-HS iteration method is convergent. Moreover, we present the optimal parameter θ* of the LME-HS method and discuss the spectral properties of corresponding preconditioned matrix. Finally, the numerical experiments are used to verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

In this paper, we mainly propose some new preconditioners for a special class of 3 × 3 block saddle point problems, which can arise from the time-dependent Maxwell equations and some other practical problems. Firstly, the solvability of this kind of problem is investigated under suitable assumptions. Then, we analyze the corresponding eigenvalues of the new preconditioned matrices presented in this work. Furthermore, we show that proposed preconditioned matrices only have at most three distinct eigenvalues. Finally, two numerical examples are also carried to verify the effectiveness of proposed preconditioners. [ABSTRACT FROM AUTHOR]

Synthetic aperture radar (SAR) image classification is a fundamental research in the interpretation of SAR images. The previous methods are unilaterally based on statistical features or spatial features, which cannot capture features with complete SAR image characteristics and unavoidably limits the performance for classification. In this article, novel sample weighting and class adversarial training strategies are proposed to fuse complementary SAR characteristics. Based on these, a distribution and structure match auxiliary classifier generative adversarial network (DSM-ACGAN) is constructed for high-quality discriminative feature learning. Particularly, the characteristics of statistical distribution and spatial structure are jointly considered in class adversarial training of DSM-ACGAN. On the one hand, DSM-ACGAN sets the true SAR image characteristics as goals for the generator to learn generative models of each category. On the other hand, and more importantly, it guides the discriminator to simultaneously capture the desired statistical and structural features. Through the class adversarial processing, the discriminative feature learning progressively improves and contributes to classification. Additionally, class-balanced and plausible samples can be generated. Experimental results on three broad SAR images from different satellites confirm the effectiveness of class adversarial training and the superiority of discriminative feature learning in DSM-ACGAN. Visual performance and quantitative metrics also show the state-of-the-art performance of the novel model. [ABSTRACT FROM AUTHOR]

Polarimetric synthetic aperture radar (PolSAR) as a typical multi-channel sensor can obtain refined geometrical and geophysical information. In the PolSAR land cover classification task, feature extraction is regarded as a critical step for the final classification. It can employ multi-modal features from the original scattering data, polarimetric target decomposition, and other transformation space. Then, how to efficiently combine multi-modal polarimetric information and extract discriminant features is an important challenge for PolSAR image processing. Graph embedding methods have become a significant technique to deal with feature extraction and dimensionality reduction (DR) problems in recent years. It provides a unified linearization framework in machine learning and other pattern recognition tasks. In this article, an extended tensor embedding framework is introduced to extract the intrinsic features for PolSAR land cover classification. First, each pixel is represented by a feature cube that is constructed by groups of polarimetric scattering signals and target decomposition features in a fixed size patch. Second, an intrinsic matrix is constructed to describe the original geometrical and statistical properties of the samples, and a penalty matrix is designed to represent some constraints. Third, the vector-based algorithms are transformed into tensor space in an unified framework and based on the pair of matrices to obtain the projection matrices in each mode by an iterative optimization process. The effectiveness of the proposed methods is demonstrated on three RADARSAT2 data sets covering the regions of Xi’an, San Francisco, and Flevoland, respectively. The visualization and quantification results show that the proposed method has superiority in land cover classification. [ABSTRACT FROM AUTHOR]

This article proposes a novel autoencoder (AE) network based on the distribution of polarimetric synthetic aperture radar (POL-SAR) data matrix, called a mixture autoencoder (MAE). Through a detailed analysis of the data distribution POL-SAR data matrix, a normalization method is also presented in succession. The proposed MAE defines the data error term in the loss function according to the data distribution. It can be regarded as a process of unsupervised feature extraction designed specifically for POL-SAR data matrix. Then, a softmax classifier is trained with the help of data features and the corresponding label information. Next, a stacked MAE (SMAE) network is reasonably constructed by considering the data distribution among different layers. Finally, this article also presents a classification network through discarding the decoder process of the proposed SMAE and connecting with a softmax classifier. The SMAE is trained layer by layer using the unlabeled data. The softmax classifier is also trained with a small number of labeled pixels. With parameters obtained from the above-mentioned procedures as the initial parameters, the whole classification network is trained by the labeled pixels to get a well-trained model, which is used for predicting the corresponding label of the pixel in the data set. Three real POL-SAR data sets, including the AIR-SAR L-band data of Flevoland, The Netherlands, are used in the experiments. Compared with one classical algorithm and two related models with the similar structure, both the proposed methods show improvements in overall accuracy and efficiency as well as possess better adaptability of the parameter and preferable consistency with the classification performance. [ABSTRACT FROM AUTHOR]

Automatic object detection from remote sensing images has attracted a significant attention due to its importance in both military and civilian fields. However, the low confidence of the candidates restricts the recognition of potential objects, and the unreasonable predicted boxes result in false positives (FPs). To address these issues, an accurate and fast object detection method called the refined single-shot multibox detector (RSSD) is proposed, consisting of a single-shot multibox detector (SSD), a refined network (RefinedNet), and a class-specific spatial template matching (STM) module. In the training stage, fed with augmented samples in diverse variation, the SSD can efficiently extract multiscale features for object classification and location. Meanwhile, RefinedNet is trained with cropped objects from the training set to further enhance the ability to distinguish each class of objects and the background. Class-specific spatial templates are also constructed from the statistics of objects of each class to provide reliable object templates. During the test phase, RefinedNet improves the confidence of potential objects from the predicted results of SSD and suppresses that of the background, which promotes the detection rate. Furthermore, several grotesque candidates are rejected by the well-designed class-specific spatial templates, thus reducing the false alarm rate. These three parts constitute a monolithic architecture, which contributes to the detection accuracy and maintains the speed. Experiments on high-resolution panchromatic (PAN) images of satellites GaoFen-2 and JiLin-1 demonstrate the effectiveness and efficiency of the proposed modules and the whole framework. [ABSTRACT FROM AUTHOR]

In order to better interpret polarimetric synthetic aperture radar (PolSAR) images, many scholars tend to do target decomposition for PolSAR images and utilize the obtained features to perform subsequent classification. These target decomposition features play an important role in terrain classification but completely utilizing them produces a high computational complexity. Furthermore, some features have a negative impact on the classification task. Therefore, selecting the appropriate amount of high-quality features is of great significance to the classification task. In this paper, we propose a convolutional neural network (CNN)-based feature selection algorithm for PolSAR image classification. First, we design a 1-D CNN for feature selection, then train the designed network with all the decomposition features to obtain a trained model. Second, the Kullback–Leibler distance (KLD) between different features is utilized as a standard to select feature subsets. Third, feature subsets with excellent performance form the final results. Due to the special structure of the 1-D CNN, repetitively training model is avoided when the input changes. Different from traditional feature selection methods, our method considers the performance of features combination rather than single feature contribution. To this end, the feature subsets selected by the proposed method are more useful to the classification task. Innovatively introducing KLD in the selection stage avoids random selection and improves the selection efficiency. Finally, we validate the performance of selected feature subsets in traditional and deep learning classification frameworks. Experiments demonstrate that features selected by the proposed method have a good performance comparing with others on three real PolSAR data sets. [ABSTRACT FROM AUTHOR]

It is well known that various features extraction approaches are utilized in polarimetric synthetic aperture (PolSAR) terrain classification for representing the data characteristic. It needs relevant and effective feature fusion algorithms to process complicated features. To address this issue, this article presents a multimodal sparse representation (MSR) framework based algorithm to fuse the different feature vectors from the complicated data space. Polarimetric data features, decomposition features, and the texture features from Pauli colour-coded image are selected to represent multimodal data in different observation modes. The corresponding multimodal manifold regularizations are added to MSR framework to approximate the data structure. Considering the independence and correlation of features, the intrinsic affinity matrices are calculated from this framework. They are processed via local preserve projection algorithm to project the multimodal features into a low dimensionally intrinsic feature space for subsequent classification. Three datasets are utilized in experiments, Western Xi'an, Flevoland, and San Francisco Bay datasets from the Radarsat-2 system in C-band. The effect of regularization parameters and different dimensional fused features are analysed in visualization and quantitation performance. The experiment results demonstrate that the effectiveness and validity of proposed method are superior to other state-of-art methods. [ABSTRACT FROM AUTHOR]

Although traditional symbolic reasoning methods are highly interpretable, their application in knowledge graph link prediction is limited due to their low computational efficiency. In this paper, we propose a new neural symbolic reasoning method: RNNCTPs, which improves computational efficiency by re-filtering the knowledge selection of Conditional Theorem Provers (CTPs), and is less sensitive to the embedding size parameter. RNNCTPs is divided into relation selectors and predictors. The relation selectors are trained efficiently and interpretably, so that the whole model can dynamically generate knowledge for the inference of predictor. In all four datasets, this method shows competitive performance against traditional methods on the link prediction task, and can have higher applicability than CTPs on some datasets. [ABSTRACT FROM AUTHOR]

Sparse representation has been widely used in the field of remote sensing image super-resolution (SR) to restore a high-quality image from a low-resolution (LR) image, e.g., from the blurred and downsampled version of an LR image’s high-resolution (HR) counterpart. It is well known that each image patch can be represented by a linear combination of the atoms of an overcomplete dictionary, and we can obtain an expression of sparse coefficients by $l_{1}$ norm regularization. Owing to the lack of an inner relationship between image patches and an image’s global information, the traditional methods of jointly training two overcomplete dictionaries cannot obtain good SR results. Therefore, we propose an effective approach for remote sensing image SR based on sparse representation. More specifically, a novel global joint dictionary model (GJDM) is used to explore the prior knowledge of images, including local and global characteristics. First, we train two dictionaries for detail image patches and HR patches. Second, in order to enhance the inner relationship between image patches, we introduce a global self-compatibility model for global regularization. Finally, the sparse representation and the local and nonlocal constraints are integrated to improve the performance of the model, and the fast adaptive shrinkage-thresholding algorithm is employed to solve the convex optimization problem in the GJDM. Compared with other methods, the results of the proposed method show good SR performance in preserving details and texture information and significant improvement in a peak signal-to-noise ratio. [ABSTRACT FROM AUTHOR]

Sparsity-regularized synthetic aperture radar (SAR) imaging framework has shown its remarkable performance to generate a feature-enhanced high-resolution image, in which a sparsity-inducing regularizer is involved by exploiting the sparsity priors of some visual features in the underlying image. However, since the simple prior of low-level features is insufficient to describe different semantic contents in the image, this type of regularizer will be incapable of distinguishing between the target of interest and unconcerned background clutters. As a consequence, the features belonging to the target and clutters are simultaneously affected in the generated image without concerning their underlying semantic labels. To address this problem, we propose a novel semantic information guided generative framework for target-oriented SAR image formation, which aims at enhancing the interested target scatters while suppressing the background clutters. First, we develop a new semantics-specific regularizer for image formation by exploiting the statistical properties of different semantic categories in a target scene SAR image. In order to infer the semantic label for each pixel in an unsupervised way, we moreover induce a novel high-level prior-driven regularizer and some semantic causal rules from the prior knowledge. Finally, our regularized framework for image formation is further derived as a simple iteratively reweighted $\ell _{1}$ minimization problem that can be conveniently solved by many off-the-shelf solvers. Experimental results demonstrate the effectiveness and superiority of our framework for SAR image formation in terms of target enhancement and clutters suppression, compared with the state of the arts. Additionally, the proposed framework opens a new direction of devoting some machine learning strategies to image formation, which can benefit the subsequent decision-making tasks. [ABSTRACT FROM AUTHOR]

Features play an important role in the learning technologies and pattern recognition methods for polarimetric synthetic aperture (PolSAR) image interpretation. In this paper, based on the subspace clustering algorithms, we combine sparse representation, low-rank representation, and manifold graphs to investigate the intrinsic property of PolSAR data. In this algorithm framework, the features are projected through the projection matrix with the sparse or/and the low rank characteristic in the low dimensional space. Meanwhile, different kinds of manifold graphs explore the geometry structure of PolSAR data to make the projected feature more discriminative. Those learned matrices, that are constrained by the sparsity and low rank terms can search for a few points from the samples and capture the global structure. The proposed algorithms aim at constructing a projection matrix from the subspace clustering algorithms to achieve the features benefiting for the subsequent PolSAR image classification. Experiments test the different combinations of those constraints. It demonstrates that the proposed algorithms outperform other state-of-art linear and nonlinear approaches with better quantization and visualization performance in PolSAR data from spaceborne and airborne platforms. [ABSTRACT FROM AUTHOR]

The robustness of the supervised polarimetric synthetic aperture radar (PolSAR) image classification is severely affected by two main aspects, namely, the quantity and quality of the labeled training pixels. Specifically, limited manually labeled pixels with respect to the large scale of PolSAR image have limited the performance of the automatic classification methods, while manually labeled training pixels shall be unfaithful with the speckle and impure cell for their low qualities. In order to address the above two fundamental problems, we propose a robust semisupervised probability graphic-based classification framework. First, a semisupervised learning scheme is implemented to simultaneously exploit both labeled and unlabeled pixels for information compensation. Moreover, structural relationship among neighboring pixels inducing from the prior information is further benefit to reduce the influence of limited labeled pixels. Second, a robust classification loss function is added in the process of training classifier to enhance the robustness to the noisy labeled pixels. Third, unfaithful limited labeled data can be settled with a hybrid generative/discriminative classification framework, where labeled and unlabeled pixels are simultaneously exploited for learning high-level feature for the low-quality pixels. The effectiveness of the proposed framework on the specific aspect is validated in experiments on real PolSAR data sets, which reveal the superiority in both visual performance and classification accuracy compared with the state-of-the-art methods. Totally speaking, our model has improved the classification accuracy by at least 20% on data set Flevoland, 10% on Oberpfaffenhofen, and 5% on Weihe River than the compared ones. [ABSTRACT FROM PUBLISHER]

Sparse coding-inspired high-resolution inverse synthetic aperture radar (ISAR) imaging based on multistage compressive sensing (CS) is proposed in this paper. To achieve high cross-range resolution within a short coherent processing interval (CPI), we divide the ISAR imaging process into multiple stages, each of which can be regarded as sparse-coding processing, and the encoding and decoding matrices can be obtained by solving an optimization problem using the error backpropagation algorithm. The decoding matrix can be regarded as a sparse dictionary, and we can recover the ISAR image by exploiting the smoothed l_0 norm algorithm based on the decoding matrix and CS theory. The ISAR image is then transformed into a time-domain echo by the inverse fast Fourier transform, and the echo can be regarded as the input for the next stage. At each stage, the resolution is gradually improved until we obtain the expected high-resolution ISAR image. Thus, the signals' energy can be fully accumulated step by step, and we can obtain a less noise and focused ISAR image. Experimental results show that the proposed method can obtain higher quality ISAR images compared with other current techniques, and that it is an effective approach to ISAR imaging within a short CPI. [ABSTRACT FROM AUTHOR]

Restoring Poissonian noise images have drawn a lot of attention in recent years. There are many regularization methods to solve this problem and one of the most famous methods is the total variation model. In this paper, by adding a quadratic regularization on TGV regularization part, a new image restoration model is proposed based on second-order total generalized variation regularization. Then the split Bregman iteration algorithm was used to solve this new model. The experimental results show that the proposed model and algorithm can deal with Poisson image restoration problem well. What's more, the restoration model performance is significantly improved both in visual effect and objective evaluation indexes. [ABSTRACT FROM AUTHOR]

Linear systems with multiple right-hand sides arise in many applications. To solve such systems efficiently, a new deflated block GCROT( m , k ) method is explored in this paper by exploiting a modified block Arnoldi deflation. Incorporating this modified block Arnoldi deflation, the new algorithm can address the possible linear dependence at each iteration during the block Arnoldi procedure and reduces expensive computational operations. Moreover, as a block version of GCROT( m , k ), the new method inherits the property of easy operability. Finally, some numerical examples also illustrate the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

As is well known, the nonnegative matrix factorization (NMF) is a dimension reduction method that has been widely used in image processing, text compressing, signal processing, and so forth. In this paper, an algorithm on nonnegative matrix approximation is proposed. This method is mainly based on a relaxed active set and the quasi-Newton type algorithm, by using the symmetric rank-one and negative curvature direction technologies to approximate the Hessian matrix. The method improves some recent results. In addition, some numerical experiments are presented in the synthetic data, imaging processing, and text clustering. By comparing with the other six nonnegative matrix approximation methods, this method is more robust in almost all cases. [ABSTRACT FROM AUTHOR]

As is well known, the nonnegative matrix factorization (NMF) is a dimension reduction method that has been widely used in image processing, text compressing, signal processing, and so forth. In this paper, an algorithm on nonnegative matrix approximation is proposed. This method is mainly based on a relaxed active set and the quasi-Newton type algorithm, by using the symmetric rankone and negative curvature direction technologies to approximate the Hessian matrix. The method improves some recent results. In addition, some numerical experiments are presented in the synthetic data, imaging processing, and text clustering. By comparing with the other six nonnegative matrix approximation methods, this method is more robust in almost all cases. [ABSTRACT FROM AUTHOR]

Abstract: Fractional diffusion equations (FDEs) have recently been paid much attention. Finding accurate and efficient methods for solving FDEs has become an active research undertaking. In this paper, an efficient method based on the shifted Chebyshev-tau idea is presented to solve an initial-boundary value problem for the FDEs. The method is derived by expanding the required approximate solution as the elements of shifted Chebyshev polynomials. Using the operational matrix of the fractional derivative, the problem can be reduced to a set of linear algebraic equations. From a computational point of view, the solution obtained by this method is in excellent agreement with those obtained by previous work in the literature and only a small number of shifted Chebyshev polynomials is needed. [Copyright &y& Elsevier]

Abstract: Recently, a Newton’s iterative method is attracting more and more attention from various fields of science and engineering. This method is generally quadratically convergent. In this paper, some Chebyshev-type methods with the third order convergence are analyzed in detail and used to compute approximate inverse preconditioners for solving the linear system Ax = b. Theoretic analysis and numerical experiments show that Chebyshev’s method is more effective than Newton’s one in the case of constructing approximate inverse preconditioners. [Copyright &y& Elsevier]

This paper provides a novel pointwise-adaptive speckle filter based on local homogeneous-region segmentation with pixel-relativity measurement. A ratio distance is proposed to measure the distance between two speckled-image patches. The theoretical proofs indicate that the ratio distance is valid for multiplicative speckle, while the traditional Euclidean distance failed in this case. The probability density function of the ratio distance is deduced to map the distance into a relativity value. This new relativity-measurement method is free of parameter setting and more functional compared with the Gaussian kernel-projection-based ones. The new measurement method is successfully applied to segment a local shape-adaptive homogeneous region for each pixel, and a simplified strategy for the segmentation implementation is given in this paper. After segmentation, the maximum likelihood rule is introduced to estimate the true signal within every homogeneous region. A novel evaluation metric of edge-preservation degree based on ratio of average is also provided for more precise quantitative assessment. The visual and numerical experimental results show that the proposed filter outperforms the existing state-of-the-art despeckling filters. [ABSTRACT FROM AUTHOR]

Abstract: In this work, the sign distribution for all inverse elements of general tridiagonal H-matrices is presented. In addition, some computable upper and lower bounds for the entries of the inverses of diagonally dominant tridiagonal matrices are obtained. Based on the sign distribution, these bounds greatly improve some well-known results due to Ostrowski (1952) , Shivakumar and Ji (1996) , Nabben (1999) and recently given by Peluso and Politi (2001) , Peluso and Popolizio (2008) and so forth. It is also stated that the inverse of a general tridiagonal matrix may be described by parameters ( and ) instead of ones as given by El-Mikkawy (2004) , El-Mikkawy and Karawia (2006) and Huang and McColl (1997) . According to these results, a new symbolic algorithm for finding the inverse of a tridiagonal matrix without imposing any restrictive conditions is presented, which improves some recent results. Finally, several applications to the preconditioning technology, the numerical solution of differential equations and the birth–death processes together with numerical tests are given. [Copyright &y& Elsevier]

Abstract: In this study, some upper and lower bounds for singular values of a general complex matrix are investigated, according to singularity and Wielandt’s lemma of matrices. Especially, some relationships between the singular values of the matrix and its block norm matrix are established. Based on these relationships, one may obtain the effective estimates for the singular values of large matrices by using the lower dimension norm matrices. In addition, a small error in Piazza (2002) [G. Piazza, T. Politi, An upper bound for the condition number of a matrix in spectral norm, J. Comput. Appl. Math. 143 (1) (2002) 141–144] is also corrected. Some numerical experiments on saddle point problems show that these results are simple and sharp under suitable conditions. [Copyright &y& Elsevier]

Some new upper and lower bounds on determinants are presented for diagonally dominant matrices and general H-matrices by using different methods. These bounds are some improvements of results given by Ostrowski (1952) and (1937), Price (1951), Wang and Zhang (2002), Huang and Liu (2005), and so forth. In addition, these bounds are also used to localize some numerical characters (e.g., the minimum eigenvalues, singular values and condition numbers) of certain matrices. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, the following inclusion sets are under certain conditions presented for singular values of a matrix :andwhere and for any , ρ(B) and denote the Perron root of the nonnegative matrix satisfying for all with and the set of all nontrivial circuits γ in Γ(A)—the associated directed digraph of A, respectively. Subsequently, relations between these two results are investigated. Theoretic analysis and numerical examples show that these estimates are more precise than recent corresponding results in some cases. [Copyright &y& Elsevier]

A matrix with positive row sums and all its off-diagonal elements bounded above by their corresponding row averages is called a B-matrix by J. M. Peña in References (SIAM J. Matrix Anal. Appl. 2001; 22:1027–1037) and (Numer. Math. 2003; 95:337–345). In this paper, it is generalized to more extended matrices—MB-matrices, which is proved to be a subclass of the class of P-matrices. Subsequently, we establish relationships between defined and some already known subclasses of P-matrices (see, References SIAM J. Matrix Anal. Appl. 2000; 21:67–78; Linear Algebra Appl. 2004; 393:353–364; Linear Algebra Appl. 1995; 225:117–123). As an application, some subclasses of P-matrices are used to localize the real eigenvalues of a real matrix. Copyright © 2007 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

In this paper, some optimal inclusion intervals of matrix singular values are discussed in the set ΩA of matrices equimodular with matrix A. These intervals can be obtained by extensions of the Gerschgorin-type theorem for singular values, based only on the use of positive scale vectors and their intersections. Theoretic analysis and numerical examples show that upper bounds of these intervals are optimal in some cases and lower bounds may be non-trivial (i.e. positive) when PA is a H-matrix, where P is a permutation matrix, which improves the conjecture in Reference (Linear Algebra Appl 1984; 56:105-119). Copyright © 2006 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]

Abstract: For the Hadamard product A ∘ A −1 of an M-matrix A and its inverse A −1, we give new lower bounds for the minimum eigenvalue of A ∘ A −1. These bounds are strong enough to prove the conjecture of Fiedler and Markham [An inequality for the Hadamard product of an M-matrix and inverse M-matrix, Linear Algebra Appl. 101 (1988) 1–8]. [Copyright &y& Elsevier]

Abstract: A new criterion for identifying generalized diagonally dominant matrices (GDDM, or -matrices) was given in [K. Ojiro, H. Niki, M. Usui, A note on a new criterion for the -matrix property, J. Comput. Appl. Math. 150 (2003) 293–302], where the criterion is effective when the given matrix is irreducible. However, when the matrix is reducible, it may lead to a false answer. In this work, to overcome this drawback, we present an improved version. The new method is always convergent and needs fewer iterations than the earlier one. An interesting sufficient and necessary condition for any weakly diagonally dominant matrix (WDDM) to be an -matrix is obtained. Finally, some numerical examples for the effectiveness of the proposed algorithm are presented. [Copyright &y& Elsevier]

Abstract: In this paper, some improvements on [Int. J. Comput. Math. 80 (June) (2003) 799–803] are presented for a bound for moduli of eigenvalues of iterative matrices. And subsequently, we investigate the relationships among them and show that the new bounds are better than the earlier ones [Int. J. Comput. Math. 53 (1994) 203–217; J. Comput. Math. 7 (2) (1986) 41–46; Int. J. Comput. Math. 80 (June) (2003) 799–803] under some conditions. [Copyright &y& Elsevier]

Abstract: Some improvements of a Ky Fan theorem are presented for matrix eigenvalue inclusion problems. Subsequently, we investigate the relationships between К 0(A), the union of Ky Fan theorem sets, S(A), the union of the Theorem 2.1 sets, and К 1(A), the union of the Theorem 2.2 sets, for eigenvalue inclusions of an n × n complex matrix A. If σ(A) denotes the spectrum of A, we show that is valid for any weakly irreducible n × n complex matrix A with n ≥ 2. [Copyright &y& Elsevier]

Ship detection is a challenging task for synthetic aperture radar (SAR) images. Ships have arbitrary directionality and multiple scales in SAR images. Furthermore, there is a lot of clutter near the ships. Traditional detection algorithms are not robust to these situations and easily cause redundancy in the detection area. With the continuous improvement in resolution, the traditional algorithms cannot achieve high-precision ship detection in SAR images. An increasing number of deep learning algorithms have been applied to SAR ship detection. In this study, a new ship detection network, known as the instance segmentation assisted ship detection network (ISASDNet), is presented. ISASDNet is a two-stage detection network with two branches. A branch is called an object branch and can extract object-level information to obtain positioning bounding boxes and classification results. Another branch called the pixel branch can be utilized for instance segmentation. In the pixel branch, the designed global relational inference layer maps the features to interaction space to learn the relationship between ship and background. The global reasoning module (GRM) based on global relational inference layers can better extract the instance segmentation results of ships. A mask assisted ship detection module (MASDM) is behind the two branches. The MASDM can improve detection results by interacting with the outputs of the two branches. In addition, a strategy is designed to extract the mask of SAR ships, which enables ISASDNet to perform object detection training and instance segmentation training at the same time. Experiments carried out two different datasets demonstrated the superiority of ISASDNet over other networks. [ABSTRACT FROM AUTHOR]

The optimal parameters of polarimetric scattering decomposition are critical to classify the pixels in polarimetric synthetic aperture radar (PolSAR) images by utilizing the method of machine learning. Therefore, span-based mutual information (Sp-MI) is proposed to lighten the dependence on labeling information, and then a heuristic representative learning scheme is also given by artificial neural network (ANN) to classify parameters separately with the increasing sequence according to the values of Sp-MI. Furthermore, an innovative method of using the sine function is presented to map the parameters of angular, and a min-max scaling method is applied to complete the procedure of normalization. Except for the support vector machine, three ANN-based classifiers are implemented to verify the rationality and effectiveness of the proposed representative learning and normalization scheme. Meanwhile, the classification method is compared with four similar comparison methods on three real PolSAR images. Finally, the classification results show the effectiveness of the proposed Sp-MI and the validation of the representative learning scheme in the aspect of classification overall accuracy and visual effect. [ABSTRACT FROM AUTHOR]

Efficient activation and the conversion of methane to value‐added products have long attracted interest. Chemical looping reforming has become a competitive technology to transform methane into syngas that can act as raw material for various chemicals. In this paper, the methane chemical looping reforming reactivity of the Ce−Co composite oxygen carriers fabricated with different preparation methods, including solution combustion, ball‐milling, hydrothermal, and co‐precipitation method, are investigated on a fixed‐bed. Various bulk/surface characterizations (TEM, XRD, BET, XPS, Raman, etc.) were applied to reveal the effect of preparation methods on the properties of these oxygen carriers. The results indicate that the oxygen carrier prepared by the solution combustion method exhibits excellent methane conversion, high syngas yield, and desirable H2/CO ratio, as well as stable regenerability because of its relatively small crystal size, large specific surface area, high oxygen vacancy concentration and Ce3+/Ce4+ ratio. This work may be helpful in the preparation of highly active and stable oxygen carriers for chemical looping reforming reaction of methane. [ABSTRACT FROM AUTHOR]