Search

Two algorithms have been proposed in this paper. One is another scanning t test of trend change points in regression slope coefficients in two phases, along with a coherency analysis of trend changes between two time series. It is different from the previously published scanning Fmax test of trend changes. The second is a fuzzy weighted moving average (FWMA). Then, the algorithms were applied to two series of monthly temperatures over global land and ocean surfaces for 1850–2021. The applied results show that significant changes in segment trends appeared in two gradations on the interdecadal and intradecadal scales. All subsample regression models were found to fit well with those data. Global warming started in April 1976 with a good coherency of warming trends between land and sea. The global warming "hiatus" mainly occurred in SST cooling from November 2001 to April 2008, but was not evinced over land on interdecadal scales. The "land/sea warming contrast" was detected only in their anomaly series, but disappeared in their standardized differences. We might refer to the anomalies in distribution N(0,s) as "perceptual" indicators and refer to the standardized differences in N(0,1) as "net" indexes. [ABSTRACT FROM AUTHOR]

Pixel‐based generative image inpainting has been widely researched over recent years and certain level of success via deep learning of feature representations and hallucinations of missing pixel values from surrounding backgrounds have also been reported in the literature. However, existing approaches rely on context‐based attentions and progressive inferences to capture the pixel correlations yet such pixel‐based approaches often fail to adapt to the constantly varying ranges and distances among surrounding background pixels. On the other hand, the modelling cost is also increasingly expensive whenever correlations of those pixels at longer distance away are to be exploited. To resolve the problem, we implement the principle of learning and hallucinating frequency components rather than pixel values. Therefore, we can avoid the dilemma that, on one hand the wish is to exploit all correlated pixels inside the image no matter how far away they are spatially located, but on the other, the price of increasing the modelling cost incurred by those pixels far away from the missing regions has to be paid. Extensive experiments carried out verify the effectiveness of the proposed method, which outperforms the representative existing state of the arts in terms of all assessment metrics. [ABSTRACT FROM AUTHOR]

Domain adaptation techniques learn transferable knowledge from a source domain to a target domain and train models that generalize well in the target domain. Unfortunately, a majority of the existing techniques are only applicable to scenarios that the target-domain data in the task of interest is available for training, yet this is not often true in practice. In general, human beings are experts in generalization across domains. For example, a baby can easily identify the bear from a clipart image after learning this category of animal from the photo images. To reduce the gap between the generalization ability of human and that of machines, we propose a new solution to the challenging zero-shot domain adaptation (ZSDA) problem, where only a single source domain is available and the target domain for the task of interest is not accessible. Inspired by the observation that the knowledge about domain correlation can improve our generalization ability, we explore the correlation between source domain and target domain in an irrelevant knowledge task ($ \mathbb {K}$ K -task), where dual-domain samples are available. We denote the task of interest as the question task ($ \mathbb {Q}$ Q -task) and synthesize its non-accessible target-domain as such that these two tasks have the shared domain correlation. In order to realize our idea, we introduce a new network structure, i.e., conditional coupled generative adversarial networks (CoCoGAN), by extending the coupled generative adversarial networks (CoGAN) into a conditioning model. With a pair of coupling GANs, our CoCoGAN is able to capture the joint distribution of data samples across two domains and two tasks. For CoCoGAN training in a ZSDA task, we introduce three supervisory signals, i.e., semantic relationship consistency across domains, global representation alignment across tasks, and alignment consistency across domains. Experimental results demonstrate that our method can learn a suitable model for the non-accessible target domain and outperforms the existing state of the arts in both image classification and semantic segmentation. [ABSTRACT FROM AUTHOR]

As the global facial mask market continues to grow, consumers have put forward higher requirements for the functionality and ingredients of mask products. Ordinary facial masks mostly use ordinary non-woven fabrics as the mask base fabric and are used with essence. Preservatives are generally added. At the same time, they are susceptible to the influence of the external environment and are easily oxidized, causing the mask to deteriorate and cause skin allergic reactions. In addition, traditional facial masks have problems such as poor fit with the skin, poor breathability, insufficient absorption of nutrient solutions, and easy dripping. The high specific surface area and high porosity of a nanofiber mask prepared by electrospinning technology are beneficial to the skin's absorption of nutrients, and it has good fit with the skin and strong breathability. A unique advantage of this nanofiber mask is that it uses spray. After the mask is sprayed with water or essence, the water-soluble polymer within it can be quickly dissolved, saving a lot of time. Nanofiber facial mask products can effectively solve consumer pain points and are conducive to the high-end development of facial masks. Therefore, this article combines needleless electrospinning technology to develop a new solid-state, preservative-free, quick-dissolving nanofiber facial mask that can be prepared on a large scale. Based on needleless electrospinning technology, this article deeply explores the process parameters and their influencing mechanisms for preparing nanofiber, quick-dissolving facial masks to achieve the stable preparation of nanofiber facial masks with the best morphology; a comprehensive analysis of the structure and influence of nanofiber facial masks from micro and macro perspectives demonstrates their performance and allows evaluation of them. The experimental results show that the mask morphology is optimal under the process conditions of using a spinning liquid of 20% collagen peptide solution, a spinning voltage of 30 kV, a collection distance of 19 cm, and a liquid supply speed of 130 mL/h. [ABSTRACT FROM AUTHOR]

Supervised deep learning techniques have achieved success in many computer vision tasks. However, most deep learning methods are data hungry and rely on a large number of labeled data in the training process. This work introduces an unsupervised deep clustering framework and studies the discovery of knowledge from a set of unlabeled data samples. Specifically, we propose a new network structure for both representation learning and GMM (Gaussian Mixture Model)-based representation modeling. In the training process of our proposed network, we not only adjust the Gaussian components to better model the distribution of representations, but also adjust the data representations towards their associating Gaussian centers to provide more adaptive support for the GMM. In this way, we take the data representations as the supervisory signal for the update of the GMM parameters and the GMM as the supervisory signal for the update of the representations, yet keeping the entire deep clustering as unsupervised. Consequently, we train the network based on an objective function with two learning targets. With the first target, we learn a GMM to model the representations properly and make each Gaussian component to be compact as much as possible. With the second target, we improve the inter-cluster distance by adapting the cluster centers to be further away from their neighbors. Thus, the training procedure simultaneously improves the intra-cluster compactness and inter-cluster separability for all the evolved clusters. Experimental results on eight datasets show that the proposed method can improve the clustering performance in comparison with the existing state of the art techniques. [ABSTRACT FROM AUTHOR]

• Based on spectral analysis, we propose a novel deep learning framework, SA-Net for deep image clustering. • Our proposed SA-Net can learn deep features and reveal deep correlations among data samples. • SA-Net integrates different spectral analysis procedures to achieve better clustering performances. • SA-Net identifies local similarities among images and is robust against occlusions. Although supervised deep representation learning has attracted enormous attentions across areas of pattern recognition and computer vision, little progress has been made towards unsupervised deep representation learning for image clustering. In this paper, we propose a deep spectral analysis network for unsupervised representation learning and image clustering. While spectral analysis is established with solid theoretical foundations and has been widely applied to unsupervised data mining, its essential weakness lies in the fact that it is difficult to construct a proper affinity matrix and determine the involving Laplacian matrix for a given dataset. In this paper, we propose a SA-Net to overcome these weaknesses and achieve improved image clustering by extending the spectral analysis procedure into a deep learning framework with multiple layers. The SA-Net has the capability to learn deep representations and reveal deep correlations among data samples. Compared with the existing spectral analysis, the SA-Net achieves two advantages: (i) Given the fact that one spectral analysis procedure can only deal with one subset of the given dataset, our proposed SA-Net elegantly integrates multiple parallel and consecutive spectral analysis procedures together to enable interactive learning across different units towards a coordinated clustering model; (ii) Our SA-Net can identify the local similarities among different images at patch level and hence achieves a higher level of robustness against occlusions. Extensive experiments on a number of popular datasets support that our proposed SA-Net outperforms 11 benchmarks across a number of image clustering applications. [ABSTRACT FROM AUTHOR]

Steganographer detection is to detect culprit users, who attempt to hide confidential information with steganography, among many innocent users. By incorporating the knowledge of true embedding probability map that illustrates the probability distribution of embedding messages in the corresponding image, content-adaptive steganography and steganalysis have made great progress. Unfortunately, true embedding probability map is inappropriate for steganographer detection method due to the significant challenges that the steganographic algorithm and the embedding payload are usually unknown in the task of steganographer detection. In this paper, we propose a novel content-adaptive selective steganographer detection method incorporated with learning-based embedding probability estimation. The embedding probability estimation is first formulated as a pixel-wise segmentation and recognition problem and is integrated into multi-class dilated residual learning model to extract the discriminative features. In the end, the steganographer is identified by local factor outlier with the selective strategy. Extensive experiments demonstrate that the estimated embedding probability map shows robustness against different steganographic algorithms and different payloads. From our experiments, we also find that the proposed content-adaptive selective steganographer detection framework integrated by the estimated embedding probability map achieves low detection error rates in both spatial and frequency domains. [ABSTRACT FROM AUTHOR]

While the recently reported deep photo style transfer [1] has shown improved results in photographic style transfer, it is found sensitive to spatial differences in the semantic segmentation of the inputs when applied to head portraits. The stylized image could incur ghost shadows when the segmented regions between the input image and the style reference image are spatially different. To minimize such a risk and reduce the influence of spatial differences between the input image and the style image, we introduce a spatial transformation strategy in this paper before style transfer. By maximizing the normalized cross-correlation of the feature maps, we propose to apply a series of affine transformations to the reference image and use those spatially-transformed images as the style reference to achieve a robust style transfer, when the semantically segmented regions of both the input image and the reference image are used as inputs to the pre-trained convolutional neural network. Consequently, the incurred ghost shadows can be minimized or eliminated. Experiments show that the proposed can perform well even when the semantic segmentation of the two images have large spatial differences, achieving significant level of robustness compared with the existing benchmark [1]. [ABSTRACT FROM AUTHOR]

This article, as part I, introduces three algorithms and applies them to both series of the monthly stream flow and rainfall in Xijiang River, southern China. The three algorithms include (1) normalization of probability distribution, (2) scanning U test for change points in correlation between two time series, and (3) scanning F-test for change points in variances. The normalization algorithm adopts the quantile method to normalize data from a non-normal into the normal probability distribution. The scanning U test and F-test have three common features: grafting the classical statistics onto the wavelet algorithm, adding corrections for independence into each statistic criteria at given confidence respectively, and being almost objective and automatic detection on multiscale time scales. In addition, the coherency analyses between two series are also carried out for changes in variance. The application results show that the changes of the monthly discharge are still controlled by natural precipitation variations in Xijiang's fluvial system. Human activities disturbed the ecological balance perhaps in certain content and in shorter spells but did not violate the natural relationships of correlation and variance changes so far. [ABSTRACT FROM AUTHOR]

This article, as part II, illustrates applications of other two algorithms, i.e., the scanning F test of change points in trend and the scanning t test of change points in mean, to both series of the normalized streamflow index (NSI) at Makou section in the Xijiang River and the normalized precipitation index (NPI) over the watershed of Xijiang River. The results from these two tests show mainly positive coherency of changes between the NSI and NPI. However, some minor negative coherency patches may expose somewhat impacts of human activities, but they were often associated with nearly normal climate periods. These suggest that the runoff still depends upon well the precipitation in the Xijiang catchment. The anthropogenic disturbances have not yet reached up to violating natural relationship on the whole in this river. [ABSTRACT FROM AUTHOR]

To improve the performance of ensemble techniques for temporal data clustering, we propose a novel bi-weighted ensemble in this paper to solve the initialization and automated model selection problems encountered by all HMM-based clustering techniques and their applications. Our proposed ensemble features in a bi-weighting scheme in the process of examining each partition and optimizing consensus function on these input partitions in accordance with their level of importance. Within our proposed scheme, the multiple partitions, generated by HMM-based K-models under different initializations, are optimally re-consolidated into a representation of bi-weighted hypergraph, and the final consensus partition is generated and optimized via the agglomerative clustering algorithm in association with a dendrogram-based similarity partitioning (DSPA). In comparison with the existing state of the arts, our proposed approach not only achieves the advantage that the number of clusters can be automatically determined, but also the superior clustering performances on a range of temporal datasets, including synthetic dataset, time series benchmark, and real-world motion trajectory datasets. [ABSTRACT FROM AUTHOR]

While existing logistic regression suffers from overfitting and often fails in considering structural information, we propose a novel matrix-based logistic regression to overcome the weakness. In the proposed method, 2D matrices are directly used to learn two groups of parameter vectors along each dimension without vectorization, which allows the proposed method to fully exploit the underlying structural information embedded inside the 2D matrices. Further, we add a joint ℓ2,1-norm on two parameter matrices, which are organized by aligning each group of parameter vectors in columns. This added co-regularization term has two roles--enhancing the effect of regularization and optimizing the rank during the learning process. With our proposed fast iterative solution, we carried out extensive experiments. The results show that in comparison to both the traditional tensor-based methods and the vector-based regression methods, our proposed solution achieves better performance for matrix data classifications. [ABSTRACT FROM AUTHOR]

Existing vector-based multimedia classification often incurs loss of space-time information and requires generation of high-dimensional vectors. To explore a possible new solution for the problem, we propose a novel tensor-based logistic regression algorithm via Tucker decomposition to complete multimedia classification. In order to strengthen the classification process, ℓ F -norm is used for regularization term. A logistic Tucker regression model is established to achieve effective extraction of principal components out of the tensors, and hence reduce the dimension of inputs to improve the efficiency of multimedia classification. To evaluate the proposed algorithm, we carried out extensive experiments on a number of data sets, including two second-order grayscale image datasets and one third-order video sequence dataset. All the results indicate that our proposed algorithm outperforms the existing state-of-the-arts in relevant areas. [ABSTRACT FROM AUTHOR]

Modern Supervisory Control and Data Acquisition (SCADA) systems used by the electric utility industry to monitor and control electric power generation, transmission and distribution are recognized today as critical components of the electric power delivery infrastructure. SCADA systems are large, complex and incorporate increasing numbers of widely distributed components. The presence of a real time intrusion detection mechanism, which can cope with different types of attacks, is of great importance in order to defend a system against cyber attacks. This defense mechanism must be distributed, cheap and above all accurate, since false positive alarms or mistakes regarding the origin of the intrusion mean severe costs for the system. Recently an integrated detection mechanism, namely IT-OCSVM , was proposed, which is distributed in a SCADA network as a part of a distributed intrusion detection system (DIDS), providing accurate data about the origin and the time of an intrusion. In this paper we also analyze the architecture of the integrated detection mechanism and we perform extensive simulations based on real cyber attacks in a small SCADA testbed in order to evaluate the performance of the proposed mechanism. [ABSTRACT FROM AUTHOR]

Spike (S) protein, a homotrimeric glycoprotein, is the most important antigen target for SARS-CoV-2 vaccines. A complete simulation of the advanced structure of this homotrimer during subunit vaccine development is the most likely method to improve its immunoprotective effects. In this study, preparation strategies for the S protein receptor-binding domain, S1 region, and ectodomain trimer nanoparticles were designed using ferritin nanoparticle self-assembly technology. The Bombyx mori baculovirus expression system was used to prepare three nanoparticle vaccines with high expression levels recorded in silkworms. The results in mice showed that the nanoparticle vaccine prepared using this strategy could induce immune responses when administered via both the subcutaneous administration and oral routes. Given the stability of these ferritin-based nanoparticle vaccines, an easy-to-use and low-cost oral immunization strategy can be employed in vaccine blind areas attributed to shortages of ultralow-temperature equipment and medical resources in underdeveloped areas. Oral vaccines are also promising candidates for limiting the spread of SARS-CoV-2 in domestic and farmed animals, especially in stray and wild animals. [Display omitted] • Nanoparticle Spike antigen expressed in fusion with ferritin has stronger immunogenicity. • Both intramuscular and oral immunizations induce potent immune responses. • Antibodies induced by nanoparticle antigens have neutralizing effect on both origin strain and delta variants of SARS-CoV-2. [ABSTRACT FROM AUTHOR]

Abstract: As one of the most important cultural heritages, ink and wash paintings (IWPs) play an important role in the world of traditional Chinese arts. In comparison with western arts, the Chinese IWPs have the unique feature that the art form is primarily populated with limited number of content elements, such as stones, mountains, flowers, and animals etc. and hence most likely different art pieces share similar content, making it difficult to differentiate in terms of content alone. In this paper, we propose to extract histogram-based local feature and global feature to characterize different aspects of art styles, and such features are applied to drive neural networks to complete the classification of IWPs in terms of individual artistic descriptors. We then propose a windowed and entropy balanced fusion scheme to make integrated decisions to optimize the final classification and recognition results. Extensive evaluation via experiments is also reported, which supports that the proposed algorithm achieves good performances, outperforming the existing benchmark techniques and hence providing an excellent potential for computerized analysis and management of traditional Chinese IWPs. [Copyright &y& Elsevier]

Abstract: This study applies the scanning t -test to the tree-ring chronology of Scots pine (Pinus sylvestris L.) from Finnish Lapland for 7638 years for detecting significant changes of subseries means. Analyses are done on multi-centennial scales and the results are demonstrated both as figures and tables. Firstly, the tree-ring data were processed to smooth the high frequency fluctuations by using an 81-year low-pass Gaussian filter following the Maximum Entropy Spectrum analysis, which revealed two significant periodicities on 80.3–87.7-year and 30.8–31.8-year scales. Secondly, the scanning t -test detected 20 change-points in the low-passed series of the tree-ring chronology. The 7638-year tree-ring chronology was partitioned into 21 episodes with comparatively high or low subseries-mean index. The episode durations range from 105 to 845 years. On longer temporal scales, a decreasing tree-ring trend (indicating palaeoclimatic cooling) appeared during the period from 4960BC to 2442BC, an increasing trend (indicating palaeoclimatic warming) occurred during years 2441BC to 449BC, then a roughly normal stage followed between 425BC and 1148AD. The last 850 years contain three episodes of obvious changes: the episode in 1149–1447AD in high index, the episode in 1448AD–1735AD as a low index, and the episode in 1736–2004AD showing back to high index again. Thirdly, all the detected 21 episodes were verified by comparing the results with the previously published studies and their interpretation. This verification was expanded to some works on the ice-core and glacier in the Greenland and the Alps respectively, in the regional view. [Copyright &y& Elsevier]

Abstract: Temporal data have many distinct characteristics, including high dimensionality, complex time dependency, and large volume, all of which make the temporal data clustering more challenging than conventional static datasets. In this paper, we propose a HMM-based partitioning ensemble based on hierarchical clustering refinement to solve the problems of initialization and model selection for temporal data clustering. Our approach results four major benefits, which can be highlighted as: (i) the model initialization problem is solved by associating the ensemble technique; (ii) the appropriate cluster number can be automatically determined by applying proposed consensus function on the multiple partitions obtained from the target dataset during clustering ensemble phase; (iii) no parameter re-estimation is required for the new merged pair of cluster, which significantly reduces the computing cost of its final refinement process based on HMM agglomerative clustering; and finally (iv) the composite model is better in characterizing the complex structure of clusters. Our approach has been evaluated on synthetic data and time series benchmark, and yields promising results for clustering tasks. [Copyright &y& Elsevier]

• A deep learning method is proposed using different optimizers and transfer learning to classify Pneumonia patients. • Preparing a dataset of around 5300 X-ray images for pneumonia detection. • The proposed deep transfer learning method is trained on a benchmark open dataset of chest x-ray images. • Presenting the optimization results, precision, recall, accuracy, and F1-score for proposed method. • In proposed method achieve better accuracy in the detection rate than other techniques. Pneumonia is a disease that leads to the death of individuals within a short period since the flow of fluid in the lungs. Hence, initial diagnosis and drugs are very important to avoid the progress of the disease. This paper proposes a novel deep learning approach for automatic detection of pneumonia using deep transfer learning to simplify the detection process with improved accuracy. This work was aimed to preprocess the input chest X-ray images to identify the presence of pneumonia using U-Net architecture based segmentation and classifies the pneumonia as normal and abnormal (Bacteria, viral) using pre-trained on ImageNet dataset models such as ResNet50, InceptionV3, InceptionResNetV2. Besides, to extract the efficient features and improve accuracy of pre-trained models two optimizers, namely, Adam and Stochastic Gradient Descent (SGD) used and its performances are analyzed with batch sizes of 16 and 32. Based on the values obtained, the performances of undertaken pre-trained models are analyzed and compared with other Convolutional Neural Network (CNN) models such as DenseNet-169+SVM, VGG16, RetinaNet + Mask RCNN, VGG16 and Xception, Fully connected RCNN, etc using various measures. From the results observed that the proposed ResNet50 model work achieved 93.06% accuracy, 88.97 % precision rate, 96.78% Recall rate and 92.71% F1-score rate, which than is higher than the other models aforementioned. [ABSTRACT FROM AUTHOR]

Abstract: Pattern recognition using Dynamic Bayesian Networks (DBNs) is currently a growing area of study. The classification performance greatly relies on the choice of a DBN model that will best describe the dependencies in each class of data. In this paper, we present DBN models trained for the classification of handwritten digit. Two approaches to improve the suitability of the models are presented. One uses a fixed DBN structure, and is based on an Evolutionary Algorithm optimisation of the selection and of the layout of the observations for each class of data. The second approach is about learning part of the structure of the models from the training set of each class. Parameter learning is then performed for each DBN. Classification results are presented for the described models, and compared with previously published results. Both approaches were found to improve the recognition rate compared to previous results. [Copyright &y& Elsevier]

In this paper, we report recent research activities under the integrated project, Live Staging of Media Events (LIVE), which is funded under European Framework-6 Programme, and illustrate how a new LIVE TV broadcasting and content production concept can be introduced to improve the existing TV broadcasting services. In comparison with existing TV content production technologies, we show case that LIVE TV broadcasting format could achieve a range of significant advantages, which can be highlighted as: (i) real-time interaction between TV content production team and viewers to ensure the best possible entertainment experiences and to allow viewers not only to view but also to participate, to influence and to control; (ii) the traditional concept of TV content director can be changed to a TV content conductor, where live TV broadcasting can be conducted on real-time bases and existing content materials can be conducted into multi-streams of TV broadcasting, providing variety of choices to allow the audience to make their own preferences in watching TV programs; (iii) introduction of significant intelligence in terms of content analysis and behavior analysis etc. to further improve the quality of services based on the fundamental concept of LIVE project. When tested under field trials for 2008 Olympic Games by ORF, the developed LIVE system has illustrated great potential as well as significant promises to improve the TV broadcasting services not only in terms of the interaction formats, but also in terms of viewing experiences as well as the style of content productions and preparations. [ABSTRACT FROM PUBLISHER]

This correspondence presents an iterative method based upon k-nearest neighbors (k-NN) regression to improve the performance of statistical feature extraction for texture image retrieval. The idea exploits the fact that an ideal feature extraction system would extract similar signatures from images characterized by the same texture and different signatures from dissimilar textures. Under the assumption that conventional statistical feature extraction contributes to sufficiently good retrieval performance, the signatures of k retrieved textures are used to update the signature of the query image using the k-NN regression algorithm. Extensive experiments show significant improvements with respect to retrieval performance in comparison to conventional statistical feature extraction. [ABSTRACT FROM AUTHOR]

Abstract: In this paper, we describe a novel multiclass boosting algorithm, EDBoost, to achieve robust face recognition directly in JPEG compressed domain. In comparison with existing boosting algorithms, the proposed EDBoost exploits Euclidean distance (ED) to eliminate non-effective weak classifiers in each iteration of the boosted learning, and hence improves both feature selection and classifier learning by using fewer weak classifiers and producing lower error rates. When applied to face recognition, the EDBoost algorithm is capable of selecting the most discriminative DCT features directly in JPEG compressed domain to achieve high recognition performances. In addition, a new DC replacement scheme is also proposed to reduce the effect of illumination changes. In comparison with the existing techniques, the proposed scheme achieves robust face recognition without losing the important information carried by all DC coefficients. Extensive experiments support the conclusion that the proposed algorithm outperforms all representative existing techniques in terms of boosted learning, multiclass classification, lighting effect reduction and face recognition rates. [Copyright &y& Elsevier]

Abstract: In this paper, we present a novel method for content adaptation and video summarization fully implemented in compressed-domain. Firstly, summarization of generic videos is modeled as the process of extracted human objects under various activities/events. Accordingly, frames are classified into five categories via fuzzy decision including shot changes (cut and gradual transitions), motion activities (camera motion and object motion) and others by using two inter-frame measurements. Secondly, human objects are detected using Haar-like features. With the detected human objects and attained frame categories, activity levels for each frame are determined to adapt with video contents. Continuous frames belonging to same category are grouped to form one activity entry as content of interest (COI) which will convert the original video into a series of activities. An overall adjustable quota is used to control the size of generated summarization for efficient streaming purpose. Upon this quota, the frames selected for summarization are determined by evenly sampling the accumulated activity levels for content adaptation. Quantitative evaluations have proved the effectiveness and efficiency of our proposed approach, which provides a more flexible and general solution for this topic as domain-specific tasks such as accurate recognition of objects can be avoided. [Copyright &y& Elsevier]

Manifold learning is an effective dimension reduction method to extract nonlinear structures from high dimensional data. Recently, manifold learning has received much attention within the research communities of image analysis, computer vision and document data analysis. In this paper, we propose a boosted manifold learning algorithm towards automatic 2D face recognition by using AdaBoost to select the best possible discriminating projection for manifold learning to exploit the strength of both techniques. Experimental results support that the proposed algorithm improves over existing benchmarks in terms of stability and recognition precision rates. [ABSTRACT FROM AUTHOR]

Abstract: A rainfall index since 1873 was constructed for the middle-lower reaches of Yangtze River basin in the eastern Asia summer monsoon (May–August) based on observations and the 5 grades of dryness/wetness in eastern China. Using data from 4 stations: Shanghai (no missing data), Nanjing, Hangzhou and Wuhan (missing data for years before 1905 and from 1938 to 1945), the sums of rainfall observations for the 4 months May–August at each station for years before 1951 was standardized. Secondly, linear regression analysis using data from Shanghai was used to infer the missing data from the other three stations. Thirdly, regressive results were corrected using the 5 grads of dryness/wetness at each station. Finally, average values were calculated over the 4 stations. For years 1951––2006, standardizations at each of the 4 stations and average over the 4 stations were calculated based on full observations. The homogeneity and change points of mean (the first moment) and of variance (the second moment) in this rainfall index series were detected by applying the scanning t-test and scanning F-test separately. The results show good homogeneity in both of mean and of variance for the two periods 1873–1950 and 1951–2006. Meanwhile, 8 climatic change points in subseries means and in subseries variances were detected on decadal time scales. This index correlates well to the rainfall variation over the middle-lower reaches and south of the Yangtze River basin based on the observational data in 1951–2006. [Copyright &y& Elsevier]

Abstract: As most of MPEG and JPEG compression standards are DCT-based, we propose a simple, low-cost and fast algorithm to extract dominant colour features directly in DCT domain without involving full decompression to access the pixel data. As dominant colour is one of the colour features used in constructing MPEG-7 dominant colour descriptors, the proposed algorithm would provide useful technique for those already compressed videos and images, where MPEG-7 dominant colour descriptors are needed. While the proposed algorithm presents advantages in terms of computing efficiency, i.e. eliminating the need of IDCT for compressed videos and images, extensive experiments also support that the proposed algorithm achieves competitive performances in extracting dominant colour features when compared with the pixel domain extraction described in MPEG-7. [Copyright &y& Elsevier]

Abstract: Video temporal segmentation is normally the first and important step for content-based video applications. Many features including the pixel difference, colour histogram, motion, and edge information etc. have been widely used and reported in the literature to detect shot cuts inside videos. Although existing research on shot cut detection is active and extensive, it still remains a challenge to achieve accurate detection of all types of shot boundaries with one single algorithm. In this paper, we propose a fuzzy logic approach to integrate hybrid features for detecting shot boundaries inside general videos. The fuzzy logic approach contains two processing modes, where one is dedicated to detection of abrupt shot cuts including those short dissolved shots, and the other for detection of gradual shot cuts. These two modes are unified by a mode-selector to decide which mode the scheme should work on in order to achieve the best possible detection performances. By using the publicly available test data set from Carleton University, extensive experiments were carried out and the test results illustrate that the proposed algorithm outperforms the representative existing algorithms in terms of the precision and recall rates. [Copyright &y& Elsevier]

Abstract: In this paper, we describe a combined approach with fuzzy logic and structure tensor towards improved enhancement of possible MCs (microcalcifications) in digital mammograms. The proposed contrast enhancement algorithm has two operational components. One is structure tensor operator and the other is fuzzy enhancement operator, both of which are arranged in parallel to process the input digital mammograms. While the structure tensor operator processes the digital mammograms and produces a corresponding eigen-image to highlight the region-of-interests, the fuzzy enhancement operator fuzzifies the mammogram via the maximum fuzzy entropy principle in fuzzy domain. As a result, the local fuzzy contrast can be extracted and modified adaptively in accordance with the information provided by the eigen-image, and those non-MCs regions are suppressed by being taken as noise. After that, the mammogram is transformed back to the pixel domain and the enhanced mammogram is constructed. Extensive experimental results show that our proposed algorithm outperforms the existing benchmark in terms of cost figures across the whole range of true positive fractions (TPF). [Copyright &y& Elsevier]

As existing video processing technology is primarily developed in the pixel domain yet digital video is stored in compressed format, any application of those techniques to compressed videos would require decompression. For discrete cosine transform (DCT)-based MPEG compressed videos, the computing cost of standard row-by-mw and column-by-column inverse DCT (LDCT) transforms for a block of 8 × 8 elements requires 4096 multiplications and 4032 additions, although practical implementation only requires 1024 multiplications and 896 additions. In this paper, we propose a new algorithm to extract videos directly from MPEG compressed domain (DCT domain) without full LDCT, which is described in three extraction schemes: 1) video extraction in 2 × 2 blocks with four coefficients; 2) video extraction in 4 x 4 blocks with four DCT coefficients; and 3) video extraction in 4 × 4 blocks with nine DCT coefficients. The computing cost incurred only requires 8 additions and no multiplication for the first scheme, 2 multiplication and 28 additions for the second scheme, and 47 additions (no multiplication) for the third scheme. Extensive experiments were carried out, and the results reveal that: 1) the extracted video maintains competitive quality in terms of visual perception and inspection and 2) the extracted videos preserve the content well in comparison with those fully decompressed ones in terms of histogram measurement. As a result, the proposed algorithm will provide useful tools in bridging the gap between pixel domain and compressed domain to facilitate content analysis with low latency and high efficiency such as those applications in surveillance videos, interactive multimedia, and image processing. [ABSTRACT FROM AUTHOR]

To improve efficiency of compressed image retrieval, we propose a novel statistical feature extraction algorithm in this paper to characterize the image content directly in its compressed domain. The statistical feature extracted is mainly through computing a set of moments directly from DCT coefficients without involving full decompression or inverse DCT. Following the algorithm design, a content-based image retrieval system is implemented especially targeting retrieving joint picture expert group compressed images. Theoretical analysis and experimental results support that the system is robust to translation, rotation and scale transform with minor disturbance, and the system achieves good performances in terms of retrieval efficiency and effectiveness. [Copyright &y& Elsevier]

Presents information on a study which proposed a content adaptive information loss distribution scheme for designing a low-cost and rate-controllable image codec. Description of a content adaptive information loss distribution scheme; Low-cost approach for rate control design; Conclusions.

Generative learning in pixel domain has achieved great success in exploiting their correlations in processing images towards desired objectives, yet learning in frequency domain could provide added benefits in exploiting pixel correlations without worrying about their spatial locations and increasing their modeling costs. In this paper, we analyze the spectral bias from a frequency perspective to overcome such limitations and hence propose a dynamic self-adaptive optimization on GAN-based generative learning, leading to a dynamic and generative logo synthesis in DCT domain. To achieve exploitation of all the pixel correlations inside the whole image regardless of their spatial locations, we introduce an approximated DCT transformation and decompose both the input images and the generated images into relatively independent DCT frequency bands. As a result, a new channel of DCT domain generative learning can be established to support the existing pixel domain learning towards improved logo synthesis. Since learning across different frequency band constantly varies, we further propose a dynamic optimization scheme to maximize the effectiveness of contributions from each individual DCT frequency band. Extensive experiments are carried out and the results in comparison with the existing state of the arts illustrate that our proposed achieves significant superiority in terms of both synthesized logo quality, integrity and variety. [ABSTRACT FROM AUTHOR]

Objectives: The public health system faces major challenges due to the double burden of diabetes mellitus (DM) and tuberculosis (TB) in China. We aimed to investigate the prevalence and impact of diabetes on patients with TB. Methods: Stratified cluster sampling was used to select 13 counties as study sites in the Zhejiang province. Patients who visited designated TB hospitals in these areas participated in this study between 1 January 2017 and 28 February 2019. Multiple logistic regression models were performed to investigate the association between DM and bacteriological and imaging results. A decision tree was used to predict the bacteriology and imaging results under the influence of DM. Results: Of 5920 patients with newly diagnosed pulmonary tuberculosis, 643 (12.16%) had DM. Patients with pulmonary TB and DM were more likely to have pulmonary cavities (adjusted odds ratio [aOR], 2.81; 95% confidence intervals [95% CI]: 2.35–3.37) and higher rates of positive bacteriological tests (aOR, 2.32; 95% CI:1.87–2.87). Decision‐tree analysis showed similar results. Conclusions: Concurrence of DM and pulmonary TB makes patients more likely to have positive bacteriological results and pulmonary cavities. Therefore, appropriate measures are necessary to promptly identify and manage patients with TB and DM. [ABSTRACT FROM AUTHOR]

Intrusion detection in real‐time systems is a problem without a profound solution. In supervisory control and data acquisition (SCADA) systems the absence of a defence mechanism that can cope with different types of intrusions is of great importance. False positive alarms or mistakes regarding the origin of the intrusion mean severe costs for the system. An integrated one‐class support vector machine (OCSVM) mechanism that is distributed in a SCADA network is presented, as a part of an intrusion detection system, providing accurate information about the origin and the time of an intrusion. The module reads the network traffic, splits traffic according to the source of the packets and creates a cluster of OCSVM models. These trained models run in parallel and can accurately and fast recognise different types of attacks. [ABSTRACT FROM AUTHOR]

Background: As a physiological signal, EEG data cannot be subjectively changed or hidden. Compared with other physiological signals, EEG signals are directly related to human cortical activities with excellent temporal resolution. After the rapid development of machine learning and artificial intelligence, the analysis and calculation of EEGs has made great progress, leading to a significant boost in performances for content understanding and pattern recognition of brain activities across the areas of both neural science and computer vision. While such an enormous advance has attracted wide range of interests among relevant research communities, EEG-based classification of brain activities evoked by images still demands efforts for further improvement with respect to its accuracy, generalization, and interpretation, yet some characters of human brains have been relatively unexplored.Methods: We propose a region-level stacked bi-directional deep learning framework for EEG-based image classification. Inspired by the hemispheric lateralization of human brains, we propose to extract additional information at regional level to strengthen and emphasize the differences between two hemispheres. The stacked bi-directional long short-term memories are used to capture the dynamic correlations hidden from both the past and the future to the current state in EEG sequences.Results: Extensive experiments are carried out and our results demonstrate the effectiveness of our proposed framework. Compared with the existing state-of-the-arts, our framework achieves outstanding performances in EEG-based classification of brain activities evoked by images. In addition, we find that the signals of Gamma band are not only useful for achieving good performances for EEG-based image classification, but also play a significant role in capturing relationships between the neural activations and the specific emotional states.Conclusions: Our proposed framework provides an improved solution for the problem that, given an image used to stimulate brain activities, we should be able to identify which class the stimuli image comes from by analyzing the EEG signals. The region-level information is extracted to preserve and emphasize the hemispheric lateralization for neural functions or cognitive processes of human brains. Further, stacked bi-directional LSTMs are used to capture the dynamic correlations hidden in EEG data. Extensive experiments on standard EEG-based image classification dataset validate that our framework outperforms the existing state-of-the-arts under various contexts and experimental setups. [ABSTRACT FROM AUTHOR]

Dengue imposes a heavy economic burden on families and society. We used surveillance data reported in 2019 to characterize the dengue epidemic in Zhejiang Province, China, which provided guidance for dengue prevention and control. Dengue epidemics mostly occurred in July to October. People aged 30–44 years, males, and commercial service workers were more likely to suffer from dengue. The epidemic areas were mainly in Hangzhou and Wenzhou. Meanwhile, we assessed the economic cost of dengue in the province from both family and organizational perspectives. The direct economic burden of dengue patients was estimated to be USD 405,038.25, and the indirect economic burden was USD 140,364.90, for a total economic burden of USD 543,213.00. The direct economic burden of dengue patients should be reduced by increasing the coverage and reimbursement of health insurance. Additionally, the total annual cost of dengue prevention and control for the government and organizational sectors was estimated to be USD 7075,654.83. Quantifying the dengue burden is critical for developing disease control strategies, allocating public health resources, and setting health policy priorities. [ABSTRACT FROM AUTHOR]

Background: Pulmonary tuberculosis (PTB) is a serious infectious disease, and the factors and pathways that influence final treatment outcomes are unclear. Here, we aimed to assess the factors that influence treatment outcomes in patients with PTB using a structural equation model.Methods: Participants completed a questionnaire covering demographics, understanding of PTB, psychological status, and history of medical treatment. Exploratory factor analysis and reliability testing were performed, and a structural equation model was constructed using the SPSS and Amos software.Results: A total of 251 participants were enrolled. Symptoms of depression were observed in 94.4% of participants, whereas 6% showed mild or greater anxiety. Through factor rotation, four common factors were extracted with a total variation of 66.15%. The structural equation model indicated that regular tuberculosis-related follow-up behaviour had a direct and positive effect on the final treatment outcome, with a path coefficient value of 0.20; the level of PTB understanding had a direct positive effect on the testing behaviour for PTB, with a path coefficient of 0.26; patients' psychological characteristics had a direct negative impact on regular testing behaviour, with a path coefficient of − 0.13. The psychological characteristics and level of disease understanding of patients exerted indirect effects on the treatment outcome by affecting the way patients approached tuberculosis detection behaviour.Conclusion: Interventions aimed at improving the treatment outcomes of patients with PTB should mainly focus on financial support and improvements in psychological status in addition to a greater understanding and knowledge of PTB. Furthermore, patients should be encouraged to undergo regular PTB testing during the follow up period, as this mediates the effect of other factors on treatment outcomes and also helps in achieving favourable treatment outcomes. [ABSTRACT FROM AUTHOR]

Abstract Video summarization has unprecedented importance to help us overview current ever-growing amount of video collections. In this paper, we propose a novel dynamic video summarization model based on deep learning architecture. We are the first to solve the imbalanced class distribution problem in video summarization. The over-sampling algorithm is used to balance the class distribution on training data. The novel two-stream deep architecture with the cost-sensitive learning is proposed to handle the class imbalance problem in feature learning. In the spatial stream, RGB images are used to represent the appearance of video frames, and in the temporal stream, multi-frame motion vectors with deep learning framework is firstly introduced to represent and extract temporal information of the input video. The proposed method is evaluated on two standard video summarization datasets and a standard emotional dataset. Empirical validations for video summarization demonstrate that our model achieves performance improvement over the existing and state-of-the-art methods. Moreover, the proposed method is able to highlight the video content with the active level of arousal in affective computing task. In addition, the proposed frame-based model has another advantage. It can automatically preserve the connection between consecutive frames. Although the summary is constructed based on the frame level, the final summary is comprised of informative and continuous segments instead of individual separate frames. [ABSTRACT FROM AUTHOR]

Highlights: • This work proposes a novel deep cascaded structure to exploit the DCT inter-block and spatial correlations in dual domains. • The proposed deep cascaded neural networks estimate the truncated high-frequency DCT coefficients in dual domains. • The performance of the proposed method significantly outperforms the state-of-the-art DCT up-sampling algorithms. Abstract Recent researches show that the high-frequency discrete cosine transform (DCT) coefficients can be estimated from low-frequency DCT coefficients by exploiting the spatial correlations. Hence, images coded by DCT such as JPEG/MJPEG/H.264, etc., can be down-sampled in DCT domain, where the high-frequency information can be accurately restored through image up-sampling. In this letter, we propose a novel deep neural network using the cascaded fully connected layers and convolution layers in dual domains (DCT and spatial domains), in order to restore high-frequency DCT coefficients from observed low-frequency DCT coefficients by exploiting the DCT inter-block and spatial correlations. In the proposed network, many recent techniques are adopted, including residual network in dual domains, batch normalization, denseNet, etc. Experimental results show that the proposed cascaded networks in dual domains significantly outperforms the state-of-the-art DCT up-sampling methods in terms of PSNR (0.63–2.57 dB gain), SSIM values, and subjective evaluations on standard image datasets Set5 and Set14. [ABSTRACT FROM AUTHOR]

Recent developments of virtual reality applications have accelerated the usage of cameras with wide-angle and telephoto/macro lens, which produce nonlinear radial lens distortions, such as barrel distortion and pincushion distortion. However, due to many reasons, the resolution of images with nonlinear lens distortions is often limited. In this paper, we address the image super-resolution (SR) for images with nonlinear lens distortions through the deep convolutional neutral network with residual learning, which can significantly improve the image quality before and after the camera calibration. The proposed deep learning network was trained using hundreds of simulated images and tested on real cameras with fisheye and macro lens. Experimental results show that the proposed image SR method outperforms state-of-the-art SR methods for various degrees of radial-based barrel and pincushion distortions. [ABSTRACT FROM AUTHOR]

Background: Men who have sex with men and women (MSMW) are the most important bridge population for HIV transmission. Condom use plays an important role for HIV infection. However the predictors for condom ues with females are not well characterized. Methods: This was a cross-sectional study. Participants were enrolled by four community-based organizations (CBOs) by offline (bathrooms, bars), and online (gay applications, chat room) from April to December 2019. Electronic questionnare was fulfilled after a face-to-face training led by CBOs. We identified predictors of inconsistent condom use with females by creating a risk score based on regression coefficients. We externally validated this score via an independent cross-sectional survey conducted in Zhejiang Province in 2021. A total of 917, 615 MSMW were included in analysis in 2019 and 2021, seperately. Results: Among 917 MSMW, 73.2% reported heterosexual behavior in the prior 6 months and 38.3% reported inconsistent condom use with females (ICUF) over that time. Compared with heterosexual/unsure MSMW, bisexual MSMW reported more male and female sex partners, higher proportion of inconsistent condom use with males, less commercial sex with males (p < 0.05). Four risky predictors of ICUF were identified: Duration of local residence ≦6 months; more than one male partner in the prior 6 months; inconsistent condom use with males in the prior 6 months; and never heard post-exposure prophylaxis (PEP). The proportions of respondents indicating ICUF in the low- (0), medium- (2–4) and high-risk (6–20) groups (according to our risk scoring system) were 11.7% (14/120), 26.9% (96/357), and 78.1% (125/160), respectively (Ptrend < 0.001). In the validation survey, the respective proportions of those reporting ICUF were 13.4% (15/112), 17.8% (24/185) and 87.3% (96/110) (Ptrend < 0.001). Conclusions: We developed and validated a predictive risk score for ICUF among MSMW; four factors were identified, of which inconsistent condom use with men was the most important. Risk reduction intervention programs should focus on MSM who report inconsistent condom use with males, never heard PEP, having multiple partners and living in local less than 6 months. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new tensor-based representation algorithm for image classification. The algorithm is realized by learning the parameter tensor for image tensors. One novelty is that the parameter tensor is learned according to the Tucker tensor decomposition as the multiplication of a core tensor with a group of matrices for each order, which endows that the algorithm preserved the spatial information of image. We further extend the proposed tensor algorithm to a semi-supervised framework, in order to utilize both labeled and unlabeled images. The objective function can be solved by using the alternative optimization method, where at each iteration, we solve the typical ridge regression problem to obtain the closed form solution of the parameter along the corresponding order. Experimental results of gray and color image datasets show that our method outperforms several classification approaches. In particular, we find that our method can implement a high-quality classification performance when only few labeled training samples are provided. [ABSTRACT FROM AUTHOR]

Visual tracking using correlation filters can achieve a high running speed. However, it is still a challenging problem to deal with drastic appearance changes (e.g. large and rapid scale variation of a target) during tracking. In this Letter, a novel dual‐directional scaling state predicting method based on correlation filtering is proposed. The adaptive horizontal and vertical searches are adopted to separately and accurately estimate the scale changes in different directions. The online scale to size module is designed to enhance the prediction robustness of the target appearance. Experimental comparisons with several state‐of‐the‐art object trackers show the effectiveness and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

Several recent deep network architectures tried to handle the depth estimation process as an image reconstruction problem, in order to overcome the shortfall that the ground truth depth data is not sufficiently available. The authors introduce and validate an efficient deeper network architecture for unsupervised depth estimation with an automated parameter optimisation. In addition, a hybrid appearance loss function is also proposed to improve the depth estimation accuracy and effectiveness. The authors' proposed model achieves the advantage that individual element of the loss function is weighted using normal distribution characteristics of a Gaussian model. The proposed ideas are validated on KITTI dataset achieving best reported results among recent state‐of‐the‐art methods. [ABSTRACT FROM AUTHOR]

Cholera caused by Vibrio cholerae O139 was first reported in Bangladesh and India in 1992. To determine the genomic epidemiology and origins of O139 in China, we sequenced 104 O139 isolates collected from Zhejiang Province, China, during 1994-2018 and compared them with 57 O139 genomes from other countries in Asia. Most Zhejiang isolates fell into 3 clusters (C1-C3), which probably originated in India (C1) and Thailand (C2 and C3) during the early 1990s. Different clusters harbored different antimicrobial resistance genes and IncA/C plasmids. The integrative and conjugative elements carried by Zhejiang isolates were of a new type, differing from ICEVchInd4 and SXTMO10 by single-nucleotide polymorphisms and presence of genes. Quinolone resistance-conferring mutations S85L in parC and S83I in gyrA occurred in 71.2% of the Zhejiang isolates. The ctxB copy number differed among the 3 clusters. Our findings provided new insights for prevention and control of O139 cholera . [ABSTRACT FROM AUTHOR]

In this paper, we propose a new algorithm for fast estimation of camera motion directly in MPEG compressed domain, which starts from an existing and commonly used camera motion model to characterize the relationship between corresponding points of the neighboring frames, and then exploits the MPEG motion estimation and compensation techniques to derive closed-form estimation of all the three camera motion parameters, including pan, tilt, and zoom. The proposed algorithm features in extremely low computing cost and high processing speed with 50% improvement over the representative existing algorithms. It is suitable for implementation on hardware, such as Pan-Tilt-Zoom (PTZ) video camera. Furthermore, the three separated processing units (pan, tilt, and zoom) can be integrated into one simplified Fast PTZ Processing Unit. Comparative experiments also reveal that the proposed algorithm achieves similar estimation precision for all the three camera motion parameters. [ABSTRACT FROM AUTHOR]