Recognizing arabic handwritten characters using deep learning and genetic algorithms

Automated techniques for Arabic content recognition are at a beginning period contrasted with their partners for the Latin and Chinese contents recognition. There is a bulk of handwritten Arabic archives available in libraries, data centers, historical centers, and workplaces. Digitization of these documents facilitates (1) to preserve and transfer the country’s history electronically, (2) to save the physical storage space, (3) to proper handling of the documents, and (4) to enhance the retrieval of information through the Internet and other mediums. Arabic handwritten character recognition (AHCR) systems face several challenges including the unlimited variations in human handwriting and the leakage of large and public databases. In the current study, the segmentation and recognition phases are addressed. The text segmentation challenges and a set of solutions for each challenge are presented. The convolutional neural network (CNN), deep learning approach, is used in the recognition phase. The usage of CNN leads to significant improvements across different machine learning classification algorithms. It facilitates the automatic feature extraction of images. 14 different native CNN architectures are proposed after a set of try-and-error trials. They are trained and tested on the HMBD database that contains 54,115 of the handwritten Arabic characters. Experiments are performed on the native CNN architectures and the best-reported testing accuracy is 91.96%. A transfer learning (TF) and genetic algorithm (GA) approach named “HMB-AHCR-DLGA” is suggested to optimize the training parameters and hyperparameters in the recognition phase. The pre-trained CNN models (VGG16, VGG19, and MobileNetV2) are used in the later approach. Five optimization experiments are performed and the best combinations are reported. The highest reported testing accuracy is 92.88%.

     

Fine-Grained Multivariate Time Series Anomaly Detection in IoT

Sensors produce a large amount of multivariate time series data to record the states of Internet of Things (IoT) systems. Multivariate time series timestamp anomaly detection (TSAD) can identify timestamps of attacks and malfunctions. However, it is necessary to determine which sensor or indicator is abnormal to facilitate a more detailed diagnosis, a process referred to as fine-grained anomaly detection (FGAD). Although further FGAD can be extended based on TSAD methods, existing works do not provide a quantitative evaluation, and the performance is unknown. Therefore, to tackle the FGAD problem, this paper first verifies that the TSAD methods achieve low performance when applied to the FGAD task directly because of the excessive fusion of features and the ignoring of the relationship’s dynamic changes between indicators. Accordingly, this paper proposes a multivariate time series fine-grained anomaly detection (MFGAD) framework. To avoid excessive fusion of features, MFGAD constructs two sub-models to independently identify the abnormal timestamp and abnormal indicator instead of a single model and then combines the two kinds of abnormal results to detect the fine-grained anomaly. Based on this framework, an algorithm based on Graph Attention Neural Network (GAT) and Attention Convolutional Long-Short Term Memory (A-ConvLSTM) is proposed, in which GAT learns temporal features of multiple indicators to detect abnormal timestamps and A-ConvLSTM captures the dynamic relationship between indicators to identify abnormal indicators. Extensive simulations on a real-world dataset demonstrate that the proposed algorithm can achieve a higher F1 score and hit rate than the extension of existing TSAD methods with the benefit of two independent sub-models for timestamp and indicator detection.     

Rapid real-time generation of super-resolution hyperspectral images through compressive sensing and GPU

Recently, compressive sensing (CS) has offered a new framework whereby a signal can be recovered from a small number of noisy non-adaptive samples. This is now an active area of research in many image-processing applications, especially super-resolution. CS algorithms are widely known to be computationally expensive. This paper studies a real time super-resolution reconstruction method based on the compressive sampling matching pursuit (CoSaMP) algorithm for hyperspectral images. CoSaMP is an iterative compressive sensing method based on the orthogonal matching pursuit (OMP). Multi-spectral images record enormous volumes of data that are required in practical modern remote-sensing applications. A proposed implementation based on the graphical processing unit (GPU) has been developed for CoSaMP using computed unified device architecture (CUDA) and the cuBLAS library. The CoSaMP algorithm is divided into interdependent parts with respect to complexity and potential for parallelization. The proposed implementation is evaluated in terms of reconstruction error for different state-of-the-art super-resolution methods. Various experiments were conducted using real hyperspectral images collected by Earth Observing-1 (EO-1), and experimental results demonstrate the speeding up of the proposed GPU implementation and compare it to the sequential CPU implementation and state-of-the-art techniques. The speeding up of the GPU-based implementation is up to approximately 70 times faster than the corresponding optimized CPU.     

Hierarchical TiO2/ZnO Nanostructure as Novel Non-precious Electrocatalyst for Ethanol Electrooxidation

Metal oxides have a higher chemical stability in comparison to metals,so they can be utilized as electrocatalysts if the activity could be enhanced.Besides the composition,the morphology of the nanostructures has a considerable impact on the electrocatalytic activity.In this work,zinc oxide nano branches-attached titanium dioxide nanofibers were investigated as an economic and stable catalyst for ethanol electrooxidation in the alkaline media.The introduced material has been synthesized by electrospinning process followed by hydrothermal technique.Briefly,electrospinning of colloidal solution consisting of titanium isopropoxide,poly(vinyl acetate) and zinc nanoparticles was performed to produce nanofibers embedding solid nanoparticles.In order to produce TiO_2 nanofibers containing ZnO nanoparticles,the obtained electrospun nanofiber mats were calcined in air at 600 °C.The formed ZnO nanoparticles were exploited as seeds to outgrow ZnO branches around the TiO_2 nanofibers using the hydrothermal technique at sub-critical water conditions in the presence of zinc nitrate and bis-hexamethylene triamine.The morphology of the final product,as well as the electrochemical measurements indicated that zinc nanoparticles content in the original electrospun nanofibers has a significant influence on the electrocatalytic activity as the best performance was observed with the nanofibers synthesized from electrospun solution containing 0.1 g Zn,and the corresponding current density was 37 mA/cm~2.Overall,this study paves a way to titanium dioxide to be exploited to synthesize effective and stable metal oxide-based electrocatalysts.     

双向组合楼板的模拟与分析

利用COSMOS/M2.6软件,考虑材料非线性,模拟并分析双向组合楼板的极限承载性能。通过与文献中双向组合楼板足尺试验的对比,验证所建有限元模型的正确性。研究楼板宽高比、长细比等参数的影响;考虑表面平整及剪力钉的影响;研究固定于板底翼板,垂直于波纹方向冷弯钢板厚度及其分布的影响;研究各初始参数对极限承载力,强、弱轴方向应力分布及板挠度的影响。在正常使用极限状态下,进行双向组合楼板的性能研究。与单向组合楼板比较,双向组合楼板承载力及弱轴方向的应力有较大提高。由于使用钢板和剪力钉,组合楼板挠度增大。     

From triadic closure to conference closure: the role of academic conferences in promoting scientific collaborations

An academic conference is not only a venue for publishing papers but also a nursery room for new scientific encounters. While previous research has investigated scientific collaboration mechanisms based on the triadic closure and focal closure, in this paper, we propose a new collaboration mechanism named conference closure. Conference closure means that scholars involved in a common conference may collaborate with each other in the future. We analyze the extent to which scholars will meet new collaborators from both the individual and community levels by using 22 conferences in the field of data mining extracted from DBLP digital library. Our results demonstrate the existence of conference closure and this phenomenon is more remarkable in conferences with high field rating and large scale attendees. Scholars involved in multiple conferences will encounter more collaborators from the conferences. Another interesting finding is that although most conference attendees are junior scholars with few publications, senior scholars with fruitful publications may gain more collaborations during the conference. Meanwhile, the conference closure still holds if we control the productivity homophily. Our study will shed light on evaluating the impact of a conference from the social function perspective based on the index of conference closure.     

A Fast and Effective Multiple Kernel Clustering Method on Incomplete Data

Multiple kernel clustering is an unsupervised data analysis method that has been used in various scenarios where data is easy to be collected but hard to be labeled. However, multiple kernel clustering for incomplete data is a critical yet challenging task. Although the existing absent multiple kernel clustering methods have achieved remarkable performance on this task, they may fail when data has a high value-missing rate, and they may easily fall into a local optimum. To address these problems, in this paper, we propose an absent multiple kernel clustering (AMKC) method on incomplete data. The AMKC method first clusters the initialized incomplete data. Then, it constructs a new multiple-kernel-based data space, referred to as K-space, from multiple sources to learn kernel combination coefficients. Finally, it seamlessly integrates an incomplete-kernel-imputation objective, a multiple-kernel-learning objective, and a kernel-clustering objective in order to achieve absent multiple kernel clustering. The three stages in this process are carried out simultaneously until the convergence condition is met. Experiments on six datasets with various characteristics demonstrate that the kernel imputation and clustering performance of the proposed method is significantly better than state-of-the-art competitors. Meanwhile, the proposed method gains fast convergence speed.     

Arrangement of Reactive Power Compensation Units in the Radial Distribution Network of Moscow Oblast

Distribution networks are the direct end-consumer connection part of an electric power system. With the development of technology, advent of new current-using equipment, and increasing population density, these networks are becoming increasingly more and more complex, elements more loaded, and power and voltage losses on the line more significant, which leads to violations of standard requirements. These problems are considered using the example of a section of the Moscow oblast network. The study presented in this article is part of the research project aimed to improve the quality of electric power in the distribution network by analyzing the current state and elaborating corrective steps. The measurements were taken using electric power quality analyzers at maximal and minimal loads. To overcome a high negative voltage deviation, the reactive power compensation unit is proposed. The procedure proposed in this article to solve the given problem consists of two phases. In phase one, loss sensitivity factors (LSFs) are defined and candidate nodes selected for installing the compensation units. This makes it possible to considerably reduce the area of searching for the optimization algorithm and, therefore, cut down the calculation time and improve the algorithm’s convergence. In phase two, the hybrid particle swarm technique is applied to optimally arrange the compensation units among the selected nodes, and choose their capacity. The hybrid optimization technique includes the particle swarm technique (PST) and the quasi-Newtonian algorithm applied after meeting the PST stopping criterion. The quasi-Newtonian algorithm is applied to cut down the time for executing iterations and making the PST more convergent. Numerical modelling is performed in the MATLAB software environment. The measurements in the distribution network of the Moscow oblast served to construct a design model with 111 nodes. According to the measurement results, the voltage level in the consumer coupling nodes considerably overrides the limits defined by GOST (State Standard) 32144–2013. Serious problems with electric power in the mains are connected with inadmissible values of established voltage deviation. The proposed hybrid algorithm of arranging reactive power compensation units makes it possible to reduce the losses of electric power in the mains, reduce the voltage deviation, and increase the line power factor.     

GloveSignature: A Virtual-Reality-Based System for Dynamic Signature Verification

A survey of the principal schemes in the literature suggested that a new way of addressing the problem of signature recognition be formulated in order to find a satisfactory solution for eliminating random forgeries. A fundamental problem in the field of off-line signature recognition is the lack of a pertinent shape representation or shape factor. This paper introduces a novel idea for a dynamic signature recognition system. An initial attempt is presented to demonstrate the data glove as an effective high-bandwidth data entry device for signature recognition. GloveSignature is a virtual-reality-based environment to support the signing process. The proposed approach retains the power to discriminate against forgeries. This paper extends the use of instrumented data gloves—gloves equipped with sensors for detecting finger bend and hand position and orientation for recognizing hand signatures. Several researchers have already explored the use of gloves in other application areas but using the gloves for the recognition of hand signatures has never been reported. An attempt is made in this research to explore the feasibility of using the 5th Glove in on-line signature recognition. Two hundred signatures were collected from 20 subjects, and features were extracted. We demonstrate the effectiveness of a hybrid technique, which is based on both the most discriminating eigenfeatures and the self-organizing maps (SOFMs) for signature recognition.