Search

Purpose: The goals of this study were to (a) classify speech movements of patients with amyotrophic lateral sclerosis (ALS) in presymptomatic and symptomatic phases of bulbar function decline relying solely on kinematic features of lips and jaw and (b) identify the most important measures that detect the transition between early and late bulbar changes. Method: One hundred ninety-two recordings obtained from 64 patients with ALS were considered for the analysis. Feature selection and classification algorithms were used to analyze lip and jaw movements recorded with Optotrak Certus (Northern Digital Inc.) during a sentence task. A feature set, which included 35 measures of movement range, velocity, acceleration, jerk, and area measures of lips and jaw, was used to classify sessions according to the speaking rate into presymptomatic (> 160 words per minute) and symptomatic (< 160 words per minute) groups. Results: Presymptomatic and symptomatic phases of bulbar decline were distinguished with high accuracy (87%), relying only on lip and jaw movements. The best features that allowed detecting the differences between early and later bulbar stages included cumulative path of lower lip and jaw, peak values of velocity, acceleration, and jerk of lower lip and jaw. Conclusion: The results established a relationship between facial kinematics and bulbar function decline in ALS. Considering that facial movements can be recorded by means of novel inexpensive and easy-to-use, video-based methods, this work supports the development of an automatic system for facial movement analysis to help clinicians in tracking the disease progression in ALS.

Purpose: The authors sought to determine an optimal set of flesh points on the tongue and lips for classifying speech movements. Method: The authors used electromagnetic articulographs (Carstens AG500 and NDI Wave) to record tongue and lip movements from 13 healthy talkers who articulated 8 vowels, 11 consonants, a phonetically balanced set of words, and a set of short phrases during the recording. We used a machine-learning classifier (support-vector machine) to classify the speech stimuli on the basis of articulatory movements. We then compared classification accuracies of the flesh-point combinations to determine an optimal set of sensors. Results: When data from the 4 sensors (T1: the vicinity between the tongue tip and tongue blade; T4: the tongue-body back; UL: the upper lip; and LL: the lower lip) were combined, phoneme and word classifications were most accurate and were comparable with the full set (including T2: the tongue-body front; and T3: the tongue-body front). Conclusion: We identified a 4-sensor set-that is, T1, T4, UL, LL--that yielded a classification accuracy (91%-95%) equivalent to that using all 6 sensors. These findings provide an empirical basis for selecting sensors and their locations for scientific and emerging clinical applications that incorporate articulatory movements.

Purpose: To quantify the articulatory distinctiveness of 8 major English vowels and 11 English consonants based on tongue and lip movement time series data using a data-driven approach. Method: Tongue and lip movements of 8 vowels and 11 consonants from 10 healthy talkers were collected. First, classification accuracies were obtained using 2 complementary approaches: (a) Procrustes analysis and (b) a support vector machine. Procrustes distance was then used to measure the articulatory distinctiveness among vowels and consonants. Finally, the distance (distinctiveness) matrices of different vowel pairs and consonant pairs were used to derive articulatory vowel and consonant spaces using multidimensional scaling. Results: Vowel classification accuracies of 91.67% and 89.05% and consonant classification accuracies of 91.37% and 88.94% were obtained using Procrustes analysis and a support vector machine, respectively. Articulatory vowel and consonant spaces were derived based on the pairwise Procrustes distances. Conclusions: The articulatory vowel space derived in this study resembled the long-standing descriptive articulatory vowel space defined by tongue height and advancement. The articulatory consonant space was consistent with feature-based classification of English consonants. The derived articulatory vowel and consonant spaces may have clinical implications, including serving as an objective measure of the severity of articulatory impairment.

Bryophytes, a monophyletic group comprising three major lineages, diversified soon after the terrestrialization of land plants. However, their internal phylogenetic relationships remain controversial. In this study, we reconstructed the ordinal and familial phylogeny of bryophytes using the largest plastid data set to date, including 549 taxa that represent almost all known orders and two‐thirds of families. The strongly supported phylogenetic inference enabled us to propose in mosses seven newly segregated families, that is, Baldwiniellaceae, Calyptrochaetaceae, Ctenidiaceae, Herpetineuraceae, Isodrepaniaceae, Pseudotaxiphyllaceae, and Rozeaceae, and one reduced family, that is, Climaciaceae. We also transferred the liverwort family Calyculariaceae from Fossombroniales to Pelliales. Recent advancements in molecular phylogeny have revolutionized bryophyte classification, tending to be more fragmental. Hence, we further propose a revised familial classification system for bryophytes that includes 45 orders and 142 families in mosses, 23 orders and 85 families in liverworts, and five orders and five families in hornworts. [ABSTRACT FROM AUTHOR]

Lonicerae japonicae flos (Jinyinhua, JYH) and Lonicerae flos (Shanyinhua, SYH) have high medical and economical value. Due to their similar appearance, the more expensive JYH is often adulterated with the cheaper SYH for economic gain. In this study, near-infrared hyperspectral imaging (HSI) was used to identify the geographical origins of JYH and SYH and differentiate JYH from SYH. Support vector classification (SVC) models using linear kernel function were established to achieve the research goals. For the identification of geographical origin, we explored the impact of different sample batches on classification performance. The overall testing sets classification accuracy of JYH and SYH was in the range of 60.10–85.59% and 63.35–91.67%, respectively. For species classification, the impact of sample geographical origins and sample batches on model performances was explored. The overall testing sets classification accuracy for distinguishing JYH and SYH was 98.46–100%. These results demonstrated the significant impact of sample sources on the performance of the models. Using SVC models, the important wavelengths contributing more to the classification were identified by recursive feature elimination (RFE). The results showed that HSI holds great potential for the identification of JYH and SYH, as well as their geographical origins. This technique can provide crucial technical support for the development and standardization of the Traditional Chinese Medicine industry. • Origin identification and species classification of Jinyinhua and Shanyinhua were studied by hyperspectral imaging. • Impact of sample batches on origin identification models were investigated. • Impact of sample origins and batches on species classification models were investigated. • Important wavelengths for origin identification and species classification were analyzed. [ABSTRACT FROM AUTHOR]

Hyperspectral band selection, which aims to select a small number of bands to reduce data redundancy and noisy bands, has attracted widespread attention in recent years. Many effective clustering-based band selection methods have been proposed to accomplish the band selection task and have achieved satisfying performance. However, most of the previous methods reshape the original hyperspectral images (HSIs) into a set of stretched band vectors, which ignore the spatial information of HSIs and the difference between diverse regions. To address these issues, a graph regularized spatial–spectral subspace clustering method for hyperspectral band selection is proposed in this paper, referred to as GRSC. Specifically, the proposed method adopts superpixel segmentation to preserve the spatial information of HSIs by segmenting their first principal component into diverse homogeneous regions. Then the discriminative latent features are generated from each segmented region to represent the whole band, which can mitigate the effect of noise on the band selection. Finally, a self-representation subspace clustering model and an l 2 , 1 -norm regularization are utilized to explore the spectral correlation among all bands. In addition, a similarity graph between region-aware latent features is adaptively learned to preserve the spatial structure of HSIs in the latent representation space. Extensive classification experimental results on three public datasets verify the effectiveness of GRSC over several state-of-the-art methods. The demo code of this work is publicly available at https://github.com/WangJun2023/GRSC. [ABSTRACT FROM AUTHOR]

• Graph attention network (GAT) is a promising framework for message passing on graphs. • Increasing the attention range to more than one-hop neighbors can negatively affect the performance of GAT, reflecting the over-smoothing risk of graph neural networks in general. • The key idea is to contextualize each node with its "structural fingerprint" that can automatically adjust to the local graph topology and edge connections in the neighborhood of the node. • Our approach provides a useful platform for different subspace of node features and various spatial scale of graph structures to "cross-talk" with each other through multi-head attention, which is more flexible than existing attention mechanism using a fixed, minimal spatial attention scale. Graph attention network (GAT) is a promising framework for message passing on graphs, but how to exploit rich, high-order structural information in the attention mechanism is still an open challenge. Furthermore, increasing the attention range to more than one-hop neighbors can negatively affect the performance of GAT, reflecting the over-smoothing risk of graph neural networks in general. In this paper, we propose an "adaptive structural fingerprint" model to fully exploit complex graph topology in graph attention networks. The key idea is to contextualize each node with its "structural fingerprint" that can automatically adjust to the local graph topology and edge connections in the neighborhood of the node. By doing this, structural interactions between the nodes can be evaluated more accurately and better confined to relevant neighbors, thus contributing to an improved attention mechanism and clearer cluster boundary. Furthermore, our approach provides a useful platform for different subspace of node features and various spatial scale of graph structures to "cross-talk" with each other through multi-head attention, which is more flexible than existing attention mechanism using a fixed, minimal spatial attention scale. Encouraging results are observed on a number of benchmark data sets including citation and social networks. [ABSTRACT FROM AUTHOR]

Accurate evaluation of the treatment result on X-ray images is a significant and challenging step in root canal therapy since the incorrect interpretation of the therapy results will hamper timely follow-up which is crucial to the patients’ treatment outcome. Nowadays, the evaluation is performed in a manual manner, which is time-consuming, subjective, and error-prone. In this article, we aim to automate this process by leveraging the advances in computer vision and artificial intelligence, to provide an objective and accurate method for root canal therapy result assessment. A novel anatomy-guided multi-branch Transformer (AGMB-Transformer) network is proposed, which first extracts a set of anatomy features and then uses them to guide a multi-branch Transformer network for evaluation. Specifically, we design a polynomial curve fitting segmentation strategy with the help of landmark detection to extract the anatomy features. Moreover, a branch fusion module and a multi-branch structure including our progressive Transformer and Group Multi-Head Self-Attention (GMHSA) are designed to focus on both global and local features for an accurate diagnosis. To facilitate the research, we have collected a large-scale root canal therapy evaluation dataset with 245 root canal therapy X-ray images, and the experiment results show that our AGMB-Transformer can improve the diagnosis accuracy from 57.96% to 90.20% compared with the baseline network. The proposed AGMB-Transformer can achieve a highly accurate evaluation of root canal therapy. To our best knowledge, our work is the first to perform automatic root canal therapy evaluation and has important clinical value to reduce the workload of endodontists. [ABSTRACT FROM AUTHOR]

Objective: To classify the fabellae and discuss the relationship between the classification of fabellae and the severity of knee osteoarthritis (KOA) in Chinese. Methods: From February 2019 to February 2020, 136 patients were measured and classified using three‐dimensional computed tomography (CT) reconstruction. According to the CT imaging characteristics, the fabellae were divided into five types: type I, a fabella on the lateral femoral condyle; type II, a fabella on the medial femoral condyle; type III, a fabella on the lateral femoral condyle and a fabella on the medial femoral condyle; type IV, two fabellae on the medial femoral condyle; and type V, two fabellae on the lateral femoral condyle. The severity of KOA was assessed on the Recht grade by magnetic resonance imaging (MRI). The data were analyzed with SPSS 24.0. Results: The classification of fabellae were correlated with KOA grades (χ2 = 35.026, P < 0.05). In terms of KOA grades, grade I and grade II were occupied most by fabellar type II (32, 72.8%); type II and other types showed significant statistical difference (P < 0.05). Grade I and grade II were also mainly fabellar type IV (four, 100%). Fabellar type V's biggest component was grade III and grade IV (six, 75%). Type IV and type V showed significant statistical difference (P < 0.05). Conclusion: The classification of fabellae were correlated with KOA grades. The type II may mean the lower KOA grades while type V may mean the higher KOA grades. [ABSTRACT FROM AUTHOR]

Deep learning (DL) techniques recently have attracted much attention in the synthetic aperture radar (SAR) automatic target recognition (ATR). Due to the coherent imaging pattern, SAR images inherently suffer from the speckle noise. To mitigate its influence, this letter proposes a joint convolutional neural network (J-CNN) for simultaneous despeckling and classification of SAR targets. It integrates a two-step process in the CNN framework but without the pooling operation during the despeckling phase. Then, a new loss function is introduced, and its partial derivatives with respect to weights are given for the training of J-CNN. Finally, comparative experiments with some classical network models are carried out based on synthetic SAR target images. The results demonstrate that the proposed method not only significantly outperforms other models under strong speckle noise condition but also has an efficient architecture with fewer weight parameters. [ABSTRACT FROM AUTHOR]

Although long noncoding RNAs (lncRNAs) have significant tissue specificity, their expression and variability in single cells remain unclear. Here, we developed ColorCells (http://rna.sysu.edu.cn/colorcells/), a resource for comparative analysis of lncRNAs expression, classification and functions in single-cell RNA-Seq data. ColorCells was applied to 167 913 publicly available scRNA-Seq datasets from six species, and identified a batch of cell-specific lncRNAs. These lncRNAs show surprising levels of expression variability between different cell clusters, and has the comparable cell classification ability as known marker genes. Cell-specific lncRNAs have been identified and further validated by in vitro experiments. We found that lncRNAs are typically co-expressed with the mRNAs in the same cell cluster, which can be used to uncover lncRNAs' functions. Our study emphasizes the need to uncover lncRNAs in all cell types and shows the power of lncRNAs as novel marker genes at single cell resolution. [ABSTRACT FROM AUTHOR]

Recently, convolutional neural networks (CNNs) have been widely used for synthetic aperture radar (SAR) image classification because of their powerful feature extraction ability and high performance. However, extracting discriminative features with limited training samples is still a challenge. Moreover, some samples may be image edge samples, which often contain multiple image categories, thus deteriorate classification accuracy. To address these issues, we propose a novel classification framework, named dense-connected global covariance network (DGCNet) with edge sample constraint (ESC). First, a dense-connected sub-network was designed, which can connect different convolutional layers of conventional CNN to strengthen feature propagation, encourage feature reuse, and alleviate gradient vanishing problem. Then, a global covariance pooling layer was introduced to fully exploit the second-order information of deep features and reduce the number of training parameters. Finally, an ESC strategy was integrated into DGCNet to further improve the classification performance by assigning a smaller weight to edge samples than non-edge samples during the training process. Experimental results on two datasets demonstrated that the proposed method achieves better classification results than several popular classification methods with limited training samples. [ABSTRACT FROM AUTHOR]

The increasing popularity of lean thinking has made the analysis of value-adding (VA) and non-value adding (NVA) a popular activity to eliminate waste and improve productivity. Despite the great success of the Taiichi Ohno's waste taxonomy and its adaptation and diffusion into the maintenance sector, a systematic classification of VA and NVA remains as a significant challenge for lean practitioners. The aim of this research is to develop a system to classify VA and NVA activities for lean applications in turnaround maintenance (TAM) projects. First, a comprehensive literature review was conducted on existing methods of defining and classifying value and waste. As a result of this review, an initial system to classify VA and NVA in TAM was proposed, which was then refined in three focus group studies conducted with a group of TAM participants. The improved classification system was evaluated by applying value stream mapping (VSM) in terms of the ontology effectiveness and efficiency using a sample case. The classification system contributes to an accurate classification of value, and waste and the relevant root causes in the TAM processes. The classification system provides an insight on a consolidated understanding and classification of VA and NVA in TAM projects. [ABSTRACT FROM AUTHOR]

It is a significant challenge to classify images using limited training samples. To this end, we formulate image classification as a multi-task multi-view (MTMV) learning problem and propose a novel Pareto optimization-based method to find the solution. Specifically, we first build a multi-objective multiple-relations-constrained MTMV model called M4 to formulate the procedure of image classification. This model integrates comprehensive relations so that more knowledge can be used when classifying the images. We formulate the model as a multi-objective optimization problem, which addresses conflicts between the inconsistencies of each item in the model and the limitation on the number of relations. To generate the final classifier for each image classification task, an effective Pareto optimization-based algorithm Pareto-M4 is proposed. Pareto-M4 first generates the Pareto-optimal solutions using a novel multi-objective solver MOQPSO and then obtains the final classifiers from all the Pareto-optimal solutions using a recombination procedure. Experiments on various real-world image data sets demonstrate the effectiveness of the proposed image classification method when limited training samples are given. [ABSTRACT FROM AUTHOR]

Graph-based semi-supervised classification (GSSC) takes labeled and unlabeled samples as vertices in a graph, and edge weights as the similarity between samples. Most GSSC methods handle each labeled sample as equally important in the graph, and they mainly focus on optimizing the graph to improve the performance. In fact, samples are not always evenly distributed. Labeled samples close to the decision boundary of different classes are generally more important than labeled samples far away from the boundary. To account for the different importances, we propose an approach called Weighted Samples based Semi-Supervised Classification (WS3C for short). WS3C firstly executes multiple clusterings on the dataset to explore the structure of samples and summarizes these clustering results. Second, it quantifies the hard-to-cluster index of each labeled sample with respect to other samples based on the summarized results and employs the index to weight that sample. Next, it constructs a graph whose edge weights are equal to the frequency of two samples grouped into the same clusters in multiple clusterings. After that, it performs semi-supervised classification based on the constructed graph and weighted samples. Empirical study on synthesized and real datasets demonstrates that assigning labeled samples with different weights significantly improves the accuracy than equally treating labeled samples. WS3C not only has better performance than other related comparing methods, but also is robust to the input parameters. • Weighting samples to improve the graph-based semi-supervised classification. • Weighting samples using hard-to-cluster indices derived from multiple clusterings. • Empirical study proves the effectiveness of weighting samples for classification. • WS3C outperforms other related and representative methods. [ABSTRACT FROM AUTHOR]

Background Two editions of the International Classification of Headache Disorders (ICHD) diagnostic criteria for 'Headache attributed to an intracranial endovascular procedure' have been published, in 2004 and 2013. Despite studies that have suggested that the former is not very practical, the ICHD-3 beta did not contain major changes. Moreover, so far no consensus exists regarding characteristics of headache after intracranial endovascular procedure. Thus, there is a need for sound suggestions to improve the ICHD-3 beta diagnostic criteria. Methods Using a prospective design, we identified consecutive patients with unruptured intracranial aneurysms (UIAs) with neuroendovascular treatment from January 2014 to December 2014. Results In total, 73 patients were enrolled, and 58 patients ultimately completed the 6-month follow-up. After the procedure, five of the 29 patients (17.2%) with pre-existing headache experienced marked worsening after the procedure, while seven of the 29 patients without prior headache developed new-onset headache post-procedurally. The headaches started within 24 hours, with a mean duration of 24-72 hours. The headaches were moderate to severe. The eligibility of these events to be considered headaches caused by neuroendovascular procedures according to the ICHD-3 beta diagnostic criteria for designation was far from ideal. Conclusions Most cases of markedly worsening headaches and new-onset headaches started within 24 hours and persisted longer than that specified in the ICHD-3 beta diagnostic criteria. Moreover, considering that some items are not very practical, the ICHD-3 beta diagnostic criteria should be revised in the light of recent literature reports. [ABSTRACT FROM AUTHOR]

The measurement of spacecraft electrical characteristics and multi-label classification issues are generally including a large amount of unlabeled test data processing, high-dimensional feature redundancy, time-consumed computation, and identification of slow rate. In this paper, a fuzzy c-means offline (FCM) clustering algorithm and the approximate weighted proximal support vector machine (WPSVM) online recognition approach have been proposed to reduce the feature size and improve the speed of classification of electrical characteristics in the spacecraft. In addition, the main component analysis for the complex signals based on the principal component feature extraction is used for the feature selection process. The data capture contribution approach by using thresholds is furthermore applied to resolve the selection problem of the principal component analysis (PCA), which effectively guarantees the validity and consistency of the data. Experimental results indicate that the proposed approach in this paper can obtain better fault diagnosis results of the spacecraft electrical characteristics’ data, improve the accuracy of identification, and shorten the computing time with high efficiency. [ABSTRACT FROM AUTHOR]

Neuroimaging techniques can measure structural and functional brain connectivity with unprecedented detail in vivo. This so-called brain connectome can be represented as high dimensional matrices corresponding to edge weights in graphs. After measuring the matrices of two cohorts (i.e., patients and healthy controls), one is often required to formulate computational network models for effective feature engineering to draw discriminative distinctions between the cohorts, as well as estimate the associated statistical significance. We designed a novel method to reveal the intrinsic features of functional matrices of discriminative power for group comparison. More specifically, by encouraging co-selection of edges connected to the same node, we preserved the discriminative edges to maximum extent. To reduce the false positive rate of the extracted discriminative edges, an optimization procedure was developed to evaluate the significance of these edges and remove trivial ones. We validated the proposed method using both synthetic data and real benchmarks, and compared it to \ell1 regularized logistic regression, univariate t-test and stability selection. The experimental results clearly showed that the proposed approach outperformed the three competing methods under various settings. In addition to increasing the F-measure of feature selection, our approach captured the endogenous, discriminative connectivity patterns consistent with recent findings in biomedical literature. This data-driven method paves a new avenue of enquiry into the inherent nature of network models for functional brain connectomes. [ABSTRACT FROM AUTHOR]

An electronic nose (E-nose) and an electronic tongue (E-tongue) have been used to characterise strawberry juices based on different processes (i.e. Microwave Pasteurisation, Steam Blanching, High Temperature Short Time Pasteurisation, Frozen-Thawed, and Freshly Squeezed). Quality parameters (vitamin C, pH, total soluble solid, total acid and sugar/acid ratio) were detected by traditional measuring methods. For qualitative discrimination, E-tongue system reached higher accuracy than E-nose did, and lower than the simultaneous utilisation. For quantitative prediction, E-tongue system had better performance ( R 2 ranged from 0.7294 to 0.8622 for the calibration, and R 2 ranged from 0.7362 to 0.9527 for the validation) than E-nose did ( R 2 ranged from 0.5344 to 0.8741 for the calibration, and R 2 ranged from 0.4444 to 0.7258 for the validation), and worse than the simultaneous utilisation ( R 2 ranged from 0.9627 to 0.9834 for the calibration, and R 2 ranged from 0.8553 to 0.8959 for the validation). Either E-nose or E-tongue was suited to classify strawberry juice, but was insufficient to predict the values of quality parameters (vitamin C, pH, total soluble solid, total acid, and sugar/acid ratio). This work indicates that the simultaneous utilisation of E-nose and E-tongue can discriminate processed juices and predict juice quality parameters successfully for the food industry. [ABSTRACT FROM AUTHOR]

A potentiometric electronic tongue (PE-tongue) and a voltammetric electronic tongue (VE-tongue) were used as rapid techniques to classify and predict the honey samples from different floral and geographical origins. The PE-tongue, which was named α-ASTREE, was developed by Alpha M.O.S. (Toulouse, France), and it comprises seven potentiometric chemical sensors. The VE-tongue was self-developed at Zhejiang University and comprises six metallic working sensors. Four types of honey of different floral origins (acacia, buckwheat, data, and motherwort) and four types of acacia honey of different geographical origins were classified by both multisensor systems. Multivariate statistical data analysis techniques such as principal component analysis (PCA) and discriminant function analysis (DFA) were used to classify the honey samples. Both types of electronic tongue have good potential to classify the honey samples, and the positions of the data point for the samples in the PCA score plots based on the VE-tongue were much more closely grouped. Three regression modes, principal component regression (PCR), partial least squares regression (PLSR), and least squared-support vector machines (LS-SVM), were applied for category forecasting. These regression models exhibited a clear indication of the prediction ability of the two types of electronic tongue, and a positive trend in the prediction of the floral and geographical origin of honey was found. Moreover, the performance of these regression models for predicting the four types of honey of different geographical origins by the VE-tongue is very stable. [ABSTRACT FROM AUTHOR]

Abstract: Various clustering algorithms have been developed since conventional hierarchical cluster analysis (HCA) and partitioning clustering algorithms have their own limitations and scopes of applications. However, in the area of e-nose where clustering is applied, the conventional algorithms (mostly HCA) still play a dominant role. In addition, comparison among different clustering methods or validation of clustering results was seldom mentioned. In this paper, we present a state-of-the-art clustering method – spectral clustering – and compare it with six conventional clustering methods: K-clustering (ISODATA, FCM and k-means) and HCA (single linkage, complete linkage and Ward's). Three external validation criteria – mutual information criteria (MI), precision and rand index (RI) – were used to evaluate clustering performances on three independent e-nose datasets. The spectral clustering outperforms with statistical significance (alpha=0.05) the performance of other methods, and the single linkage presents the worst (unacceptable) clustering result. In addition, the proposed approach – cluster validation criteria in combination with majority voting – in a way makes clustering a semi-supervised classification technique. Using this approach it is possible to compare clustering based semi-supervised methods with classification methods to find which method is better for discrimination of a certain e-nose dataset. [Copyright &y& Elsevier]

Facial recognition is a challenging area of research due to difficulties with robust face recognition (FR) under occlusion and sparse representation-based classification (SRC) only focusing on face global features. To solve these issues, we proposed an occluded FR method based on dictionary learning for sparse representation and sub-classifiers fusion (LSSRC), which efficiently combines local and overall characteristics of face images. First, we partitioned continuous but non-lapped blocks of the face by multi-resolution blocking. Then, for each block, SRC was used for feature extraction and face classification. We established a sub-block dictionary and conducted K-SVD dictionary learning, established sub-classifiers and determined weight. Finally, we conducted sub-classifier fusion recognition using voting rules with weight. Results using the AR and YaleA database showed that our algorithm achieved superior recognition performance to the existing sparse representation classification occluded FR method. [ABSTRACT FROM AUTHOR]

Most traditional hyperspectral image (HSI) classification methods relied on hand-crafted or shallow-based descriptors, which limits their applicability and performance. Recently, deep learning has gradually become the mainstream method of HSI classification, because it can automatically extract deep abstract features for classification. However, it remains a challenge to learn more meaningful features for HSI classification from a small training sample set. In this paper, a 3D cascaded spectral–spatial element attention network (3D-CSSEAN) is proposed to solve this issue. The 3D-CSSEAN integrates the spectral–spatial feature extraction and attention area extraction for HSI classification. Two element attention modules in the 3D-CSSEAN enable the deep network to focus on primary spectral features and meaningful spatial features. All attention modules are implemented though several simple activation operations and elementwise multiplication operations. In this way, the training parameters of the network are not added too much, which also makes the network structure suitable for small sample learning. The adopted module cascading pattern not only reduces the computational burden in the deep network but can also be easily operated via plug–expand–play. Experimental results on three public data sets show that the proposed 3D-CSSEAN achieved comparable performance with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Magnetic response of a superconductor depends on the thermodynamic stability of a vortex in the material. Here we show that the vortex stability has a close relation with the ratio of the magnetic field at the vortex core center to the thermodynamic critical field. This finding provides a new criterion for the classification of the superconductors according to their magnetic responses. [ABSTRACT FROM AUTHOR]