Search

Machine learning (ML) algorithms are increasingly used in power systems applications. One important application is the classification and localization of various types of transmission line faults. Using voltage and current measurements from phasor measurement units (PMUs), a number of useful features can be extracted, which can form the basis of a ML-based prediction of the fault type, line, and distance on the line. This paper proposes a technique to find the optimal number and placement of PMUs by performing thorough feature selection. The features are selected to maximize the accuracy of the ML classification and regression algorithms. The results show that for the IEEE 14 bus system, the use of only five PMUs is sufficient to obtain high levels of accuracy. For example, a testing accuracy of 99.0% and 97.1% can be achieved for the fault type and fault line location, respectively. As for the fault distance along the line, the testing MAE of 3.1% can be obtained along with an R 2 score of 94.4%. Adding more PMUs does not provide any additional value in terms of accuracy. [ABSTRACT FROM AUTHOR]

Accurate fault detection in high-voltage direct current (HVDC) transmission lines plays a pivotal role in enhancing operational efficiency, reducing costs, and ensuring grid reliability. This research aims to develop a cost-effective and high-performance fault detection solution for HVDC systems. The primary objective is to accurately identify and localize faults within the power system. In pursuit of this goal, the paper presents a comparative analysis of current and voltage characteristics between the rectifier and inverter sides of the HVDC transmission system and their associated alternating current (AC) counterparts under various fault conditions. Voltage and current features are extracted and optimized using a metaheuristic approach, specifically Harris Hawk's optimization method. Leveraging machine learning (ML) and artificial neural networks (ANN), this technique demonstrates its effectiveness in generating a fault locator with exceptional accuracy. With a substantial volume of data employed for learning and training, the Harris Hawks optimization method exhibits faster convergence compared to other metaheuristic methods examined in this study. The research findings are applied to simulate diverse fault types and unknown fault locations at multiple system points. Evaluating the fault detection system's effectiveness, quantified through metrics such as specificity, accuracy, F1 score, and sensitivity, yields remarkable results, with percentages of 99.01%, 98.69%, 98.64%, and 98.67%, respectively. This research underscores the critical role of accurate fault detection in HVDC systems, offering valuable insights into optimizing grid performance and reliability. [ABSTRACT FROM AUTHOR]

The problem of intrusion detection has new solutions, thanks to the widespread use of machine learning in the field of network security, but it still has a few issues at this time. Traditional machine learning techniques to intrusion detection rely on expert experience to choose features, and deep learning approaches have a low detection efficiency. In this paper, an intrusion detection model based on feature selection and improved one-dimensional convolutional neural network was proposed. This model first used the extreme gradient boosting decision tree (XGboost) algorithm to sort the preprocessed data, and then it used comparison to weed out 55 features with a higher contribution. Then, the extracted features were fed into the improved one-dimensional convolutional neural network (I1DCNN), and this network training was used to complete the final classification task. The feature selection and improved one-dimensional convolutional neural network (FS-I1DCNN) intrusion detection model not only solved the traditional machine learning method of relying on expert experience to extract features but also improved the detection efficiency of the model, reduced the training time while reducing the dimension, and increased the overall accuracy. In comparison to the I1DCNN model without feature extraction and the conventional one-dimensional convolutional neural network (1DCNN) model, the experimental results demonstrate that the FS-I1DCNN model's overall accuracy increases by 0.67% and 2.94%, respectively. Its accuracy, precision, recall, and F 1 -score were significantly better than those of the other intrusion detection models, including SVM and DBN. [ABSTRACT FROM AUTHOR]

Nowadays, the verification of handwritten signatures has become an effective research field in computer vision as well as machine learning. Signature verification is naturally formulated as a machine-learning task. This task is performed by determining if the signature is genuine or forged. Therefore, it is considered a two‐class classification issue. Since handwritten signatures are widely used in legal documents and financial transactions, it is important for researchers to select an efficient machine-learning technique for verifying these signatures and to avoid forgeries that may cause many losses to customers. So far, great outcomes have been obtained when using machine learning techniques in terms of equal error rates and calculations. This paper presents a comprehensive review of the latest studies and results in the last 10 years in the field of online and offline handwritten signature verification. More than 20 research papers were used to make a comparison between datasets, feature extraction, and classification techniques used in each system, taking into consideration the problems that occur in each. In addition, the general limitations and advantages of machine-learning techniques that are used to classify or extract signature features were summarized in the form of a table. We also present the general steps of the verification system and a list of the most considerable datasets available in online and offline fields. [ABSTRACT FROM AUTHOR]

With the continuous development of digital technology, music, as an important form of media, and its digital audio technology is also constantly developing, forcing the traditional music industry to start the road of digital transformation. What kind of method can be used to automatically retrieve music information effectively and quickly in vocal singing has become one of the current research topics that has attracted much attention. Aiming at this problem, it is of great research significance for the field of timbre feature recognition. With the in-depth research on timbre feature recognition, the research on timbre feature extraction by machine learning in vocal singing has also been gradually carried out, and its performance advantages are of great significance to solve the problem of automatic retrieval of music information. This paper aims to study the application of feature extraction algorithm based on machine learning in timbre feature extraction in vocal singing. Through the analysis and research of machine learning and feature extraction methods, it can be applied to the construction of timbre feature extraction algorithms to solve the problem of automatic retrieval of music information. This paper analyzed vocal singing, machine learning, and feature extraction, experimentally analyzed the performance of the method, and used related theoretical formulas to explain. The results have showed that the method for timbre feature extraction in the vocal singing environment was more accurate than the traditional method, the difference between the two was 24.27%, and the proportion of satisfied users was increased by 33%. It can be seen that this method can meet the needs of users for timbre feature extraction in the use of music software, and the work efficiency and user satisfaction are greatly improved. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles (UAVs), commonly known as drones, have been progressively prevalent due to their capability to operate quickly and their vast range of applications in a variety of real-world circumstances. The utilization of UAVs in precision farming has lately gained a lot of attention from the scientific community. This study addresses with the assistance of drones in the precision agricultural area. This paper makes significant contributions by analyzing communication protocols and applying them to the challenge of commanding a fleet of drones to protect crops from parasite infestations. In this research, the effectiveness of nine powerful deep neural network models is measured for the detection of plant diseases using diverse methodologies. These deep neural networks are adapted to the immediate situation using transfer learning and deep extraction of features approaches. The presented study takes into account the used pretrained deep learning model for extracting features and fine-tuning. The deep feature extraction characteristics are subsequently categorized using support vector machines (SVMs) and extreme learning machines (ELMs). For measuring performance, the precision, sensitivities, specific, and F1-score are all evaluated. Deep feature extraction and SVM/ELM classification generated better outcomes than transfer learning, according to the analysis result. Furthermore, the analysis of the various methodologies tries to assess their effectiveness and costs. The different approaches, for example, confront difficulties such as investigating the region in the shortest possible time feasible, while eliminating the same region being searched by more drones, detecting parasites, and stopping their spread by applying the appropriate number of pesticides. Simulation models are a significant aid to researchers in conducting to evaluate these technologies and creating specific tactics and coordinating procedures capable of effectively supporting farms and achieving the aim. The main objective of this paper is to compare the search techniques of two distinct methods of parasitic to identify performance. [ABSTRACT FROM AUTHOR]

In the field of mechanical and electrical equipment, the motor rolling bearing is a workpiece that is extremely prone to damage and failure. However, the traditional fault diagnosis methods cannot keep up with the development pace of the times because they need complex manual pretreatment or the support of specific expert experience and knowledge. As a rising star, the data-driven fault diagnosis methods are increasingly favored by scholars and experts at home and abroad. The convolutional neural network has been widely used because of its powerful feature extraction ability for all kinds of complex information and its outstanding research results in image processing, target tracking, target diagnosis, time-frequency analysis, and other scenes. Therefore, this paper introduces a convolutional neural network and applies it to motor-bearing fault diagnosis. Aiming at the shortcomings of fault signal and convolutional neural network, a large-scale maximum pooling strategy is proposed and optimized by wavelet transform to improve the fault diagnosis efficiency of motor bearing under high-voltage operation. Compared with other machine learning algorithms, the convolution neural network fault diagnosis model constructed in this paper not only has high accuracy (up to 0.9871) and low error (only 0.032) but also is simple to use. It provides a new way for motor bearing fault diagnosis and has very important economic and social value. [ABSTRACT FROM AUTHOR]

Facial expression recognition based on residual networks is important for technologies related to space human-robot interaction and collaboration but suffers from low accuracy and slow computation in complex network structures. To solve these problems, this paper proposes a multiscale feature fusion attention lightweight wide residual network. The network first uses an improved random erasing method to preprocess facial expression images, which improves the generalizability of the model. The use of a modified depthwise separable convolution in the feature extraction network reduces the computational effort associated with the network parameters and enhances the characterization of the extracted features through a channel shuffle operation. Then, an improved bottleneck block is used to reduce the dimensionality of the upper layer network feature map to further reduce the number of network parameters while enhancing the network feature extraction capability. Finally, an optimized multiscale feature lightweight attention mechanism module is embedded to further improve the feature extractability of the network for human facial expressions. The experimental results show that the accuracy of the model is 73.21%, 98.72%, and 95.21% on FER2013, CK+ and JAFFE, respectively, with a covariance of 10.14 M. Compared with other networks, the model proposed in this paper has faster computing speed and better accuracy at the same time. [ABSTRACT FROM AUTHOR]

This paper proposed a hybrid approach for the identification of encrypted traffic based on advanced mathematical modeling and computational intelligence. Network traffic identification is the premise and foundation of improving network management, service quality, and application security. It is also the focus of network behavior analysis, network planning and construction, network anomaly detection, and network traffic model research. With the increase in user and service requirements, many applications use encryption algorithms to encrypt traffic during data transmission. As a result, traditional traffic classification methods classify encrypted traffic on the network, which brings great difficulties and challenges to network monitoring and data mining. In our article, a nonlinear modified DBN method is proposed and applied to encrypted traffic identification. Firstly, based on Deep Belief Networks (DBN), this paper introduces the proposed Eodified Elliott (ME)-DBN model, analyzes the function image, and presents the ME-DBN learning algorithm. Secondly, this article designs an encrypted traffic recognition model based on the ME-DBN model. Feature extraction is carried out by training the ME-DBN model, and finally, classification and recognition are carried out by the classifier. The experimental results on the ISCX VPN-non-VPN database show that the MEDBN method proposed in this article can enhance the classification and recognition rate and has better robustness to encrypt traffic recognition from different software. [ABSTRACT FROM AUTHOR]

Diabetes mellitus is a common chronic noncommunicable disease, the main manifestation of which is the long-term high blood sugar level in patients due to metabolic disorders. However, due to excessive reliance on the clinical experience of ophthalmologists, our diagnosis takes a long time, and it is prone to missed diagnosis and misdiagnosis. In recent years, with the development of deep learning, its application in the auxiliary diagnosis of diabetic retinopathy has become possible. How to use the powerful feature extraction ability of deep learning algorithm to realize the mining of massive medical data is of great significance. Therefore, under the action of computer-aided technology, this paper processes and analyzes the retinal images of the fundus through traditional image processing and convolutional neural network-related methods, so as to achieve the role of assisting clinical treatment. Based on the admission records of diabetic patients after data analysis and feature processing, this paper uses an improved convolutional neural network algorithm to establish a model for predicting changes in diabetic conditions. The model can assist doctors to judge the patient's treatment effect by using it based on the case records of inpatient diagnosis and treatment and to predict the risk of readmission of inpatients after discharge. It also can help to judge the effectiveness of the treatment plan. The results of the study show that the model proposed in this paper has a lower probability of misjudging patients with poor recovery as good recovery, and the prediction is more accurate. [ABSTRACT FROM AUTHOR]

This paper proposes a new algorithm composition network from the perspective of machine learning, based on an in-depth study of related literature. At the same time, this paper examines the characteristics of music and develops a model for recognising musical emotions. Using the model's information entropy of pitch and intensity to extract the main melody track, note features are extracted from bar features. Finally, the cosine of the vector included angle is used to judge the similarity between feature vectors of several adjacent sections, allowing the music to be divided into several independent segments. The emotional model of music is used to analyze each segment's emotion. By quantifying music features, this paper classifies and quantifies music emotion based on the mapping relationship between music features and emotion. Music emotion can be accurately identified by the model. The model's emotion recognition accuracy is up to 93.78 percent, and the algorithm's recall rate is up to 96.3 percent, according to simulation results. The recognition method used in this paper has a higher recognition ability than other methods, and the emotion recognition result is more reliable. This paper can not only meet the composer's auxiliary creative needs, but it can also help intelligent music services. [ABSTRACT FROM AUTHOR]

Traditional text annotation-based video retrieval is done by manually labeling videos with text, which is inefficient and highly subjective and generally cannot accurately describe the meaning of videos. Traditional content-based video retrieval uses convolutional neural networks to extract the underlying feature information of images to build indexes and achieves similarity retrieval of video feature vectors according to certain similarity measure algorithms. In this paper, by studying the characteristics of sports videos, we propose the histogram difference method based on using transfer learning and the four-step method based on block matching for mutation detection and fading detection of video shots, respectively. By adaptive thresholding, regions with large frame difference changes are marked as candidate regions for shots, and then the shot boundaries are determined by mutation detection algorithm. Combined with the characteristics of sports video, this paper proposes a key frame extraction method based on clustering and optical flow analysis, and experimental comparison with the traditional clustering method. In addition, this paper proposes a key frame extraction algorithm based on clustering and optical flow analysis for key frame extraction of sports video. The algorithm effectively removes the redundant frames, and the extracted key frames are more representative. Through extensive experiments, the keyword fuzzy finding algorithm based on improved deep neural network and ontology semantic expansion proposed in this paper shows a more desirable retrieval performance, and it is feasible to use this method for video underlying feature extraction, annotation, and keyword finding, and one of the outstanding features of the algorithm is that it can quickly and effectively retrieve the desired video in a large number of Internet video resources, reducing the false detection rate and leakage rate while improving the fidelity, which basically meets people's daily needs. [ABSTRACT FROM AUTHOR]

This paper shows the health monitoring and assessment of a three-phase induction motor in abnormal conditions using a machine learning algorithm. The convolutional neural network (CNN) and recurrent neural network (RNN) algorithms are the prominent methods used in machine learning algorithms, and the combined method is known as the CRNN method. The abnormal conditions of a three phase-induction motor are represented by three-phase faults, line-to-ground faults, etc. The pattern of fault current is traced, and key features are extracted by the CRNN algorithm. The performance parameters like THD (%), accuracy, and reliability of abnormal conditions are measured with the CRNN algorithm. The assessment of abnormal conditions is being realized at the terminals of a three-phase induction motor. A fuzzy logic controller (FLC) is also used to assess such abnormalities. It is observed that performance parameters are found to be better with the CRNN method in comparison to CNN, RNN, ANN, and other methods. Such a realization makes the system more compatible with abnormality recognition. [ABSTRACT FROM AUTHOR]

Damage induced by a low-velocity impact can reduce the stability and reliability of structures. In this study, a novel low-velocity impact region identification method based on the spectral peak frequency (SPF) and support vector machine (SVM) is proposed to identify the low-velocity impact regions on a steel cantilever beam. A low-velocity impact region identification system of the cantilever beam is established by applying fiber Bragg grating (FBG) sensors, and only 2 sensors are used in this system. The power spectral density functions of the impact response signal are smoothed using the linear weighting method to remove pseudospectral peak frequencies, and then, SPFs are extracted as the features. For 25 low-velocity impact regions with dimensions of 30 mm × 10 mm, the results show that the recognition rate obtained by the proposed method is 100% and the feature vector consisting of the first two SPFs with the largest amplitude has the highest recognition rate. Through the comparative study, it is found that the recognition rate of SVM is higher than that of the probabilistic neural network (PNN) and extreme learning machine (ELM) for low-velocity impact area recognition of cantilever beams. As a result, the low-velocity impact region identification method of this paper can be applied to the real-time health monitoring of cantilever beam structures. [ABSTRACT FROM AUTHOR]

The complexity of physical verification increases rapidly with fast shrinking technology nodes. Considering only design rule checking (DRC) constraints or lithography models cannot capture the side physical effects in the fabrication process well. Thus, it is desirable to consider a more general physical verification problem with various types of hotspots. In this paper, we apply machine learning which is based on pixel-based feature extraction to deal with the generalized hotspot detection problem. First, a two-dimensional discrete Fourier transformation-based pixel extraction method is proposed to alleviate the shifting effect and produce stable hotspot features. Then, a pattern-based layout scanning approach is developed to enhance the program efficiency while preserving good detection accuracy. Finally, we design two false alarm reduction strategies to effectively reduce the number of detected nonhotspots and further improve the accuracy of hotspot position. Experimental results based on the industrial benchmarks show that our algorithm outperforms three competitive works in terms of accuracy, false alarm rate, efficiency, and time. [ABSTRACT FROM AUTHOR]

There are various problems in diagnosing and treating tumor diseases in significant hospitals. The content includes misjudgement and over-surgery issues. For example, the judgment of pulmonary nodules mainly relies on artificial experience, and most of the artificial experience is too radical. This paper is mainly based on the extensive medical data of significant hospitals, extracts the diagnosis and treatment data and digital images of similar cases from the extensive database, classifies them through the deep learning of the computer, and then proposes the control mechanism and the solution of the doctor's misjudgement and excessive medical treatment. This method mainly relies on the CT and MRI digital images of various types of tumor diseases accumulated in the history of major hospitals. Based on the preliminary judgment of each diagnosis and treatment and the results of surgical and pathological examinations, the accumulation of various types of digital images from the history is used. The features are analysed and extracted, the model is built, and finally, a predictive analysis system for this type of tumor is obtained, which can predict the benign and malignant cases of currently occurring cases and avoid the limitations and instability of artificial experience greatest extent. It is proved by experiments and combined with Spearman to remove redundancy. The redundancy removal method SVM_RFE is used for dimensionality reduction. The method can timely correct the misjudgement of the doctor's experience and effectively reduce the instability of the manual, which provides a solution for solving the contradiction between doctors and patients and improving the scientific of diagnosis and treatment. [ABSTRACT FROM AUTHOR]

The application of machine learning technology to intelligent music creation has become a very important field in music creation. The main current research on music intelligent creation methods uses fixed coding steps in audio data, which lead to weak feature expression ability. Based on convolutional neural network theory, this paper proposes a deep music intelligent creation method. The model uses a convolutional recurrent neural network to generate an effective hash code, first preprocesses the music signal to obtain a Mel spectrogram, and then inputs it into a pretrained CNN to extract from its convolutional layers. The network space details and the semantic information of musical symbols are used to construct the feature map sequence using selection strategy for the feature map of each convolutional layer, so as to solve the problem of high data feature dimension and poor recognition performance. In the simulation process, the Mel cepstral coefficient method (MFCC) was used to extract the features of four different music signals, and the features that could represent each signal were extracted through the convolutional neural network, and the continuous signals were discretized and reduced. The experimental results show that the high-dimensional music data are dimensionally reduced at the data level. After the data are compressed, the correct rate of intelligent creation is as high as 98%, and the characteristic signal distortion rate is reduced to 5% below, effectively improving the algorithm performance and the ability to create music intelligently. [ABSTRACT FROM AUTHOR]

There are various problems in diagnosing and treating tumor diseases in significant hospitals. The content includes misjudgement and over-surgery issues. For example, the judgment of pulmonary nodules mainly relies on artificial experience, and most of the artificial experience is too radical. This paper is mainly based on the extensive medical data of significant hospitals, extracts the diagnosis and treatment data and digital images of similar cases from the extensive database, classifies them through the deep learning of the computer, and then proposes the control mechanism and the solution of the doctor's misjudgement and excessive medical treatment. This method mainly relies on the CT and MRI digital images of various types of tumor diseases accumulated in the history of major hospitals. Based on the preliminary judgment of each diagnosis and treatment and the results of surgical and pathological examinations, the accumulation of various types of digital images from the history is used. The features are analysed and extracted, the model is built, and finally, a predictive analysis system for this type of tumor is obtained, which can predict the benign and malignant cases of currently occurring cases and avoid the limitations and instability of artificial experience greatest extent. It is proved by experiments and combined with Spearman to remove redundancy. The redundancy removal method SVM_RFE is used for dimensionality reduction. The method can timely correct the misjudgement of the doctor's experience and effectively reduce the instability of the manual, which provides a solution for solving the contradiction between doctors and patients and improving the scientific of diagnosis and treatment. [ABSTRACT FROM AUTHOR]

Semiautomated digital creation is increasingly important in the manipulation of electronic music. How to realize the learning of local effective features of audio data is a difficult point in the current research field. Based on recurrent neural network theory, this paper designs a semiautomatic digital creation system for electronic music for digital manipulation and genre classification. The recurrent neural network improves the transmission of electronic music information between the input and output of the network by adopting dense connections consistent with DenseNet and adopts an inception-like structure for the autonomous selection of effective recursive nuclear electronic music categories. In the simulation process, the prediction method based on semiautomatic digital audio clips is also adopted, which pays more attention to the learning of local effective features of audio data, which gives the model the ability to create audio samples of different lengths and improves the model's support for creative tasks in different scenarios. It includes the determination of the number of neurons, the selection of the function of neurons, the determination of the connection method, and the specific learning algorithm rules, and then the training samples are formed. The experimental results show that the recurrent neural network exhibits powerful feature extraction ability and classification ability of music information. The 10-fold cross-validation on GTZAN dataset and ISMIR2004 dataset has obtained 88.7% and 87.68%, surpassing similar ones. The model has reached a leading level. After further use of the MSD (Million Song Dataset) dataset for pre-semiautomatic training, the model effect has been further greatly improved. The accuracy rate on the dataset has been increased to 91.0% and 89.91%, respectively, which has improved the semiautomatic number and creative advancement. [ABSTRACT FROM AUTHOR]

Multimodal tasks based on attention mechanism and language face numerous problems. Based on multimodal hierarchical attention mechanism and genetic neural network, this paper studies the application of image segmentation algorithm in data completion and 3D scene reconstruction. The algorithm refers to the process of concentrating attention that humans subjectively pay attention to and calculates the difference between each pixel in the genetic neural network test image in the color space and the average value of the target image, which solves the problem of static feature maps and dynamic feature maps of image sequences. In addition, in view of the problem that the number of attention enhancement feature extraction modules is too large and the parameters are too large, the recursive mechanism is used as the feature extraction branch, and new model parameters are not added when the network depth is increased. The simulation results show that the accuracy of the improved image saliency detection algorithm based on the attention mechanism reaches 89.7%, and the difference between the average value of the single-point pixel and the target image is reduced to 0.132, which further promotes the practicability and reliability of the image segmentation model. [ABSTRACT FROM AUTHOR]

The lungs are COVID-19's most important focus, as it induces inflammatory changes in the lungs that can lead to respiratory insufficiency. Reducing the supply of oxygen to human cells negatively impacts humans, and multiorgan failure with a high mortality rate may, in certain circumstances, occur. Radiological pulmonary evaluation is a vital part of patient therapy for the critically ill patient with COVID-19. The evaluation of radiological imagery is a specialized activity that requires a radiologist. Artificial intelligence to display radiological images is one of the essential topics. Using a deep machine learning technique to identify morphological differences in the lungs of COVID-19-infected patients could yield promising results on digital images of chest X-rays. Minor differences in digital images that are not detectable or apparent to the human eye may be detected using computer vision algorithms. This paper uses machine learning methods to diagnose COVID-19 on chest X-rays, and the findings have been very promising. The dataset includes COVID-19-enhanced X-ray images for disease detection using chest X-ray images. The data were gathered from two publicly accessible datasets. The feature extractions are done using the gray level co-occurrence matrix methods. K-nearest neighbor, support vector machine, linear discrimination analysis, naïve Bayes, and convolutional neural network methods are used for the classification of patients. According to the findings, convolutional neural networks' efficiency linked to imaging modalities with fewer human involvements outperforms other traditional machine learning approaches. [ABSTRACT FROM AUTHOR]

Hypergraph learning is a new research hotspot in the machine learning field. The performance of the hypergraph learning model depends on the quality of the hypergraph structure built by different feature extraction methods as well as its incidence matrix. However, the existing models are all hypergraph structures built based on one feature extraction method, with limited feature extraction and abstract expression ability. This paper proposed a multimodal feature fusion method, which firstly built a single modal hypergraph structure based on different feature extraction methods, and then extended the hypergraph incidence matrix and weight matrix of different modals. The extended matrices fuse the multimodal abstract feature and an expanded Markov random walk range during model learning, with stronger feature expression ability. However, the extended multimodal incidence matrix has a high scale and high computational cost. Therefore, the Laplacian matrix fusion method was proposed, which performed Laplacian matrix transformation on the incidence matrix and weight matrix of every model, respectively, and then conducted a weighted superposition on these Laplacian matrices for subsequent model training. The tests on four different types of datasets indicate that the hypergraph learning model obtained after multimodal feature fusion has a better classification performance than the single modal model. After Laplace matrix fusion, the average time can be reduced by about 40% compared with the extended incidence matrix, the classification performance can be further improved, and the index F1 can be improved by 8.4%. [ABSTRACT FROM AUTHOR]

With a view to realizing the fault diagnosis of rotating machinery effectively, an integrated health condition detection approach for rotating machinery based on refined composite multivariate multiscale amplitude-aware permutation entropy (RCmvMAAPE), max-relevance and min-redundancy (mRmR), and whale optimization algorithm-based kernel extreme learning machine (WOA-KELM) is presented in this paper. The approach contains two crucial parts: health detection and fault recognition. In health detection stage, multivariate amplitude-aware permutation entropy (mvAAPE) is proposed to detect whether there is a fault in rotating machinery. Afterward, if it is detected that there is a fault, RCmvMAAPE is employed to extract the initial fault features that represent the fault states from the multivariate vibration signals. Based on the multivariate expansion and multiscale expansion of amplitude-aware permutation entropy, RCmvMAAPE enjoys the ability to effectively extract state information on multiple scales from multichannel series, thereby overcoming the defect of information loss in traditional methods. Then, mRmR is adopted to screen the sensitive features so as to form sensitive feature vectors, which are input into the WOA-KELM classifier for fault classification. Two typical rotating machinery cases are conducted to prove the effectiveness of the raised approach. The experimental results demonstrate that mvAAPE shows excellent performance in fault detection and can effectively detect the fault of rotating machinery. Meanwhile, the feature extraction method based on RCmvMAAPE and mRmR, as well as the classifier based on WOA-KELM, shows superior performance in feature extraction and fault recognition, respectively. Compared with other fault identification methods, the raised method enjoys better performance and the average fault recognition accuracy of the two typical cases in this paper can all reach above 98%. [ABSTRACT FROM AUTHOR]

With the rapid development of the Internet, various forms of network attack have emerged, so how to detect abnormal behavior effectively and to recognize their attack categories accurately have become an important research subject in the field of cyberspace security. Recently, many hot machine learning-based approaches are applied in the Intrusion Detection System (IDS) to construct a data-driven model. The methods are beneficial to reduce the time and cost of manual detection. However, the real-time network data contain an ocean of redundant terms and noises, and some existing intrusion detection technologies have lower accuracy and inadequate ability of feature extraction. In order to solve the above problems, this paper proposes an intrusion detection method based on the Decision Tree-Recursive Feature Elimination (DT-RFE) feature in ensemble learning. We firstly propose a data processing method by the Decision Tree-Based Recursive Elimination Algorithm to select features and to reduce the feature dimension. This method eliminates the redundant and uncorrelated data from the dataset to achieve better resource utilization and to reduce time complexity. In this paper, we use the Stacking ensemble learning algorithm by combining Decision Tree (DT) with Recursive Feature Elimination (RFE) methods. Finally, a series of comparison experiments by cross-validation on the KDD CUP 99 and NSL-KDD datasets indicate that the DT-RFE and Stacking-based approach can better improve the performance of the IDS, and the accuracy for all kinds of features is higher than 99%, except in the case of U2R accuracy, which is 98%. [ABSTRACT FROM AUTHOR]

With the increasing complexity, scale, and intelligentization of modern equipment, the maintenance cost of equipment is increasing day by day. Moreover, once an unexpected major failure occurs, it will cause loss and damage to production, economy, and safety. Based on the considerations of system reliability and safety, fault prediction has gradually become a hot topic in the field of reliability. As a new branch of machine learning, deep learning realizes deep abstract feature extraction and expression of complex nonlinear relations by stacking deep neural networks and makes its methods solve bad problems in many traditional machine learning fields. The improvement and excellent results have been achieved. This article first introduces the model structure and working principle of the classic deep learning model noise reduction autoencoder and combines the feature extraction results of the experimental data of electromechanical sensor equipment and the model characteristics to analyze that this type of model can be trained using only normal samples. Under the restriction, the reason for abnormal features can also be correctly filtered. Then, in order to suppress the overfitting of training, the dropout layer is used in this article. The dropout layer will make the hidden layer nodes to be dropped with probability p according to the set probability. Because the lost nodes are random, each training is equivalent to training a new model. It achieves an effect similar to independently training the model and then superimposing it. Experiments prove that the dropout layer is very effective in solving overfitting. Finally, this paper conducts experimental verification on the improved algorithm. The results show that the improved model has a certain improvement in accuracy under the limited training algebra. Under the same training parameters, the accuracy is increased by approximately 2.44%, and the improved model has a better training effect and can be used for electromechanical effective prediction of sensor equipment failure. [ABSTRACT FROM AUTHOR]

Text readability is very important in meeting people's information needs. With the explosive growth of modern information, the measurement demand of text readability is increasing. In view of the text structure of words, sentences, and texts, a hybrid network model based on convolutional neural network is proposed to measure the readability of English texts. The traditional method of English text readability measurement relies too much on the experience of artificial experts to extract features, which limits its practicability. With the increasing variety and quantity of text readability measurement features to be extracted, it is more and more difficult to extract deep features manually, and it is easy to introduce irrelevant features or redundant features, resulting in the decline of model performance. This paper introduces the concept of hybrid network model in deep learning; constructs a hybrid network model suitable for English text readability measurement by combining convolutional neural network, bidirectional long short-term memory network, and attention mechanism network; and replaces manual automatic feature extraction by machine learning, which greatly improves the measurement efficiency and performance of text readability. [ABSTRACT FROM AUTHOR]

Traditional machine learning algorithms are susceptible to objective factors such as video quality and weather environment in the vehicle detection of Unmanned Aerial Vehicle (UAV) videos, resulting in poor detection results. A vehicle image detection method using deep learning in UAV video is proposed. The algorithm in this paper treats surveillance video as many frames of images for vehicle detection in the image. First, perform HSV (Hue-Saturation-Value) spatial brightness translation operation on the original sample to increase the adaptability to different light conditions and sample diversity. Then, the Single Shot MultiBox Detector (SSD) model framework is used as the basis for vehicle detection. In order to obtain a better feature extraction effect, focus loss is added to the basic SSD for optimization. Finally, the trained network model is used to analyze the UAV video, and the detection performance is analyzed experimentally. The results show that the vehicle detection rate of this algorithm has reached 96.49%. It can ensure that the vehicle is accurately detected from the drone video. [ABSTRACT FROM AUTHOR]

In this paper, we have proposed a novel methodology based on statistical features and different machine learning algorithms. The proposed model can be divided into three main stages, namely, preprocessing, feature extraction, and classification. In the preprocessing stage, the median filter has been used in order to remove salt-and-pepper noise because MRI images are normally affected by this type of noise, the grayscale images are also converted to RGB images in this stage. In the preprocessing stage, the histogram equalization has also been used to enhance the quality of each RGB channel. In the feature extraction stage, the three channels, namely, red, green, and blue, are extracted from the RGB images and statistical measures, namely, mean, variance, skewness, kurtosis, entropy, energy, contrast, homogeneity, and correlation, are calculated for each channel; hence, a total of 27 features, 9 for each channel, are extracted from an RGB image. After the feature extraction stage, different machine learning algorithms, such as artificial neural network, k -nearest neighbors' algorithm, decision tree, and Naïve Bayes classifiers, have been applied in the classification stage on the features extracted in the feature extraction stage. We recorded the results with all these algorithms and found that the decision tree results are better as compared to the other classification algorithms which are applied on these features. Hence, we have considered decision tree for further processing. We have also compared the results of the proposed method with some well-known algorithms in terms of simplicity and accuracy; it was noted that the proposed method outshines the existing methods. [ABSTRACT FROM AUTHOR]

With deepening internationalization, English has become an increasingly important communication tool. Because traditional English teaching has short teacher-student interaction time, lack of oral English training environment, and single learning method, the oral English teaching is not ideal, and the students' "speaking" confidence is insufficient. Aimed at addressing the exposed problems of traditional English reading teaching, this paper proposes a multimedia-based English reading teaching mode. On this basis, establish a voice recognition phoneme network grid to detect the recognition results. Secondly, the lattice is used to generate the confusion network mesh, and the acoustic posterior probability is calculated. Then, the feature vector is input into the SVM classifier for confidence mark, and finally the feature is extracted by principal component analysis. The research shows that multimedia network teaching can teach more vividly, increasing the initiative of students. At the same time, it is shown that the speech recognition confidence learning algorithm can improve the language learning system. [ABSTRACT FROM AUTHOR]

In recent years, the role of pattern recognition in systems based on human computer interaction (HCI) has spread in terms of computer vision applications and machine learning, and one of the most important of these applications is to recognize the hand gestures used in dealing with deaf people, in particular to recognize the dashed letters in surahs of the Quran. In this paper, we suggest an Arabic Alphabet Sign Language Recognition System (AArSLRS) using the vision-based approach. The proposed system consists of four stages: the stage of data processing, preprocessing of data, feature extraction, and classification. The system deals with three types of datasets: data dealing with bare hands and a dark background, data dealing with bare hands, but with a light background, and data dealing with hands wearing dark colored gloves. AArSLRS begins with obtaining an image of the alphabet gestures, then revealing the hand from the image and isolating it from the background using one of the proposed methods, after which the hand features are extracted according to the selection method used to extract them. Regarding the classification process in this system, we have used supervised learning techniques for the classification of 28-letter Arabic alphabet using 9240 images. We focused on the classification for 14 alphabetic letters that represent the first Quran surahs in the Quranic sign language (QSL). AArSLRS achieved an accuracy of 99.5% for the K-Nearest Neighbor (KNN) classifier. [ABSTRACT FROM AUTHOR]

The feature extraction of high-precision microseismic signals is an important prerequisite for multicategory recognition of microseismic signals, and it is also an important basis for intelligent sensing modules in smart mines. Aiming at the problem of unobvious feature extraction of multiclass mine microseismic signals, this paper is based on the unsupervised learning method in the deep learning method, combined with wavelet packet energy ratio and empirical modulus singular value decomposition, and proposes a method based on wavelet packet energy and empirical modulus singular value decomposition and proposes a method (M-W&E) based on wavelet packet energy and empirical modulus singular value decomposition. This method firstly performs empirical modulus singular value decomposition and wavelet packet energy ratio on the microseismic signal to construct the basic feature vector and then uses the unsupervised learning algorithm to perform the unsupervised learning method feature fusion of the basic feature vector to construct the fused feature vector. After visualization by t-SNE, various distinctions in the fusion feature vector are more obvious. After testing the fusion feature classification using SVM, it is found that the recognition rate of the new feature after feature fusion is better than that of a single wavelet packet empirical energy component and singular value of empirical modulus, which basically meets the engineering needs and is a mine microseism. The signal extraction and feature enhancement fusion of multiclass samples provide a new idea. [ABSTRACT FROM AUTHOR]

ResNet has been widely used in the field of machine learning since it was proposed. This network model is successful in extracting features from input data by superimposing multiple layers of neural networks and thus achieves high accuracy in many applications. However, the superposition of multilayer neural networks increases their computational cost. For this reason, we propose a network model compression technique that removes multiple neural network layers from ResNet without decreasing the accuracy rate. The key idea is to provide a priority term to identify the importance of each neural network layer, and then select the unimportant layers to be removed during the training process based on the priority of the neural network layers. In addition, this paper also retrains the network model to avoid the accuracy degradation caused by the deletion of network layers. Experiments demonstrate that the network size can be reduced by 24.00%–42.86% of the number of layers without reducing the classification accuracy when classification is performed on CIFAR-10/100 and ImageNet. [ABSTRACT FROM AUTHOR]

In recent years, deep learning has made good progress and has been applied to face recognition, video monitoring, image processing, and other fields. In this big data background, deep convolution neural network has also received more and more attention. In order to extract the ancient Chinese characters effectively, the paper will discuss the structure model, pool process, and network training of deep convolution neural network and compare the algorithm with the traditional machine learning algorithm. The results show that the accuracy and recall rate of the Chinese characters in the plaque of Ming Dynasty can reach the peak, 81.38% and 81.31%, respectively. When the number of training samples increases to 50, the recognition rate of MFA is 99.72%, which is much higher than other algorithms. This shows that the algorithm based on deep convolution neural network and big data analysis has excellent performance and can effectively identify the Chinese characters under different dynasties, different sample sizes, and different interference factors, which can provide a powerful reference for the extraction of ancient Chinese characters. [ABSTRACT FROM AUTHOR]

The automatic diagnosis of Alzheimer's disease plays an important role in human health, especially in its early stage. Because it is a neurodegenerative condition, Alzheimer's disease seems to have a long incubation period. Therefore, it is essential to analyze Alzheimer's symptoms at different stages. In this paper, the classification is done with several methods of machine learning consisting of K -nearest neighbor (KNN), support vector machine (SVM), decision tree (DT), linear discrimination analysis (LDA), and random forest (RF). Moreover, novel convolutional neural network (CNN) architecture is presented to diagnose Alzheimer's severity. The relationship between Alzheimer's patients' functional magnetic resonance imaging (fMRI) images and their scores on the MMSE is investigated to achieve the aim. The feature extraction is performed based on the robust multitask feature learning algorithm. The severity is also calculated based on the Mini-Mental State Examination score, including low, mild, moderate, and severe categories. Results show that the accuracy of the KNN, SVM, DT, LDA, RF, and presented CNN method is 77.5%, 85.8%, 91.7%, 79.5%, 85.1%, and 96.7%, respectively. Moreover, for the presented CNN architecture, the sensitivity of low, mild, moderate, and severe status of Alzheimer patients is 98.1%, 95.2%,89.0%, and 87.5%, respectively. Based on the findings, the presented CNN architecture classifier outperforms other methods and can diagnose the severity and stages of Alzheimer's disease with maximum accuracy. [ABSTRACT FROM AUTHOR]

Automotive intelligence has become a revolutionary trend in automotive technology. Complex road driving conditions directly affect driving safety and comfort. Therefore, by improving the recognition accuracy of road type or road adhesion coefficient, the ability of vehicles to perceive the surrounding environment will be enhanced. This will further contribute to vehicle intelligence. In this paper, considering that the process of manually extracting image features is complicated and that the extraction method is random for everyone, road surface condition identification method based on an improved ALexNet model, namely, the road surface recognition model (RSRM), is proposed. First, the ALexNet network model is pretrained on the ImageNet dataset offline. Second, the weights of the shallow network structure after training, including the convolutional layer, are saved and migrated to the proposed model. In addition, the fully connected layer fixed to the shallow network is replaced by 2 to 3, which improves the training accuracy and shortens the training time. Finally, the traditional machine learning and improved ALexNet model are compared, focusing on adaptability, prediction output, and error performance, among others. The results show that the accuracy of the proposed model is better than that of the traditional machine learning method by 10% and the ALexNet model by 3%, and it is 0.3 h faster than ALexNet in training speed. It is verified that RSRM effectively improves the network training speed and accuracy of road image recognition. [ABSTRACT FROM AUTHOR]

Aiming at the defects of wavelet transform-based feature extraction and extreme learning machine-based classification, a novel fault diagnosis method for motor bearing, based on dual tree complex wavelet transform and artificial fish swarm optimization-kernel extreme learning machine (DTCWT-AFSO-KELM), is proposed in this paper. Firstly, the dual tree complex wavelet transform instead of the discrete wavelet transform is used to decompose the motor bearing signal; then, the features with large differentiation of motor-bearing fault are extracted; finally, the states of motor bearing are classified by using artificial fish swarm optimization-kernel extreme learning machine. In order to better prove the superiority of this method, four kinds of state data of motor bearing under the conditions of 0 HP (horsepower) load, 1 HP load, 2 HP load, and 3 HP load are used to test. The experimental results indicate that the diagnosis accuracies of DTCWT-AFSO-KELM are obviously better than those of discrete wavelet transform and artificial fish swarm optimization-kernel extreme learning machine (DWT-AFSO-KELM) or discrete wavelet transform and extreme learning machine (DWT-ELM) under different loads. [ABSTRACT FROM AUTHOR]

In view of the current absence of any deep learning algorithm for shellfish identification in real contexts, an improved Faster R-CNN-based detection algorithm is proposed in this paper. It achieves multiobject recognition and localization through a second-order detection network and replaces the original feature extraction module with DenseNet, which can fuse multilevel feature information, increase network depth, and avoid the disappearance of network gradients. Meanwhile, the proposal merging strategy is improved with Soft-NMS, where an attenuation function is designed to replace the conventional NMS algorithm, thereby avoiding missed detection of adjacent or overlapping objects and enhancing the network detection accuracy under multiple objects. By constructing a real contexts shellfish dataset and conducting experimental tests on a vision recognition seafood sorting robot production line, we were able to detect the features of shellfish in different scenarios, and the detection accuracy was improved by nearly 4% compared to the original detection model, achieving a better detection accuracy. This provides favorable technical support for future quality sorting of seafood using the improved Faster R-CNN-based approach. [ABSTRACT FROM AUTHOR]

The discriminative spatial patterns (DSP) algorithm is a classical and effective feature extraction technique for decoding of voluntary finger premovements from electroencephalography (EEG). As a purely data-driven subspace learning algorithm, DSP essentially is a spatial-domain filter and does not make full use of the information in frequency domain. The paper presents multilinear discriminative spatial patterns (MDSP) to derive multiple interrelated lower dimensional discriminative subspaces of low frequency movement-related cortical potential (MRCP). Experimental results on two finger movement tasks' EEG datasets demonstrate the effectiveness of the proposed MDSP method. [ABSTRACT FROM AUTHOR]

Structural health monitoring (SHM) is a hot research topic with the main purpose of damage detection in a structure and assessing its health state. The major focus of SHM studies in recent years has been on developing vibration-based damage detection algorithms and using machine learning, especially deep learning-based approaches. Most of the deep learning-based methods proposed for damage detection in civil structures are based on supervised algorithms that require data from the healthy state and different damaged states of the structure in the training phase. As it is not usually possible to collect data from damaged states of a large civil structure, using such algorithms for these structures may be impractical. This paper proposes a new unsupervised deep learning-based method for structural damage detection based on convolutional autoencoders (CAEs). The main objective of the proposed method is to identify and quantify structural damage using a CAE network that employs raw vibration signals from the structure and is trained by the signals solely acquired from the healthy state of the structure. The CAE is chosen to take advantage of high feature extraction capability of convolution layers and at the same time use the advantages of an autoencoder as an unsupervised algorithm that does not need data from damaged states in the training phase. Applications on the two numerical models of IASC-ASCE benchmark structure and a grid structure located at the University of Central Florida, as well as the full-scale Tianjin Yonghe Bridge, prove the efficiency of the proposed algorithm in assessing the global health state of the structures and quantifying the damage. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new method for hyperspectral images (HSI) classification, aiming to take advantage of both manifold learning-based feature extraction and neural networks by stacking layers applying locality sensitive discriminant analysis (LSDA) to broad learning system (BLS). BLS has been proven to be a successful model for various machine learning tasks due to its high feature representative capacity introduced by numerous randomly mapped features. However, it also produces redundancy, which is indiscriminate and finally lowers its performance and causes heavy computing demand, especially in cases of the input data bearing high dimensionality. In our work, a manifold learning method is integrated into the BLS by inserting two LSDA layers before the input layer and output layer separate, so the spectral-spatial HSI features are fully utilized to acquire the state-of-the-art classification accuracy. The extensive experiments have shown our method's superiority. [ABSTRACT FROM AUTHOR]

Electroencephalography-(EEG-) based control is a noninvasive technique which employs brain signals to control electrical devices/circuits. Currently, the brain-computer interface (BCI) systems provide two types of signals, raw signals and logic state signals. The latter signals are used to turn on/off the devices. In this paper, the capabilities of BCI systems are explored, and a survey is conducted how to extend and enhance the reliability and accuracy of the BCI systems. A structured overview was provided which consists of the data acquisition, feature extraction, and classification algorithm methods used by different researchers in the past few years. Some classification algorithms for EEG-based BCI systems are adaptive classifiers, tensor classifiers, transfer learning approach, and deep learning, as well as some miscellaneous techniques. Based on our assessment, we generally concluded that, through adaptive classifiers, accurate results are acquired as compared to the static classification techniques. Deep learning techniques were developed to achieve the desired objectives and their real-time implementation as compared to other algorithms. [ABSTRACT FROM AUTHOR]

Electroencephalography-(EEG-) based control is a noninvasive technique which employs brain signals to control electrical devices/circuits. Currently, the brain-computer interface (BCI) systems provide two types of signals, raw signals and logic state signals. The latter signals are used to turn on/off the devices. In this paper, the capabilities of BCI systems are explored, and a survey is conducted how to extend and enhance the reliability and accuracy of the BCI systems. A structured overview was provided which consists of the data acquisition, feature extraction, and classification algorithm methods used by different researchers in the past few years. Some classification algorithms for EEG-based BCI systems are adaptive classifiers, tensor classifiers, transfer learning approach, and deep learning, as well as some miscellaneous techniques. Based on our assessment, we generally concluded that, through adaptive classifiers, accurate results are acquired as compared to the static classification techniques. Deep learning techniques were developed to achieve the desired objectives and their real-time implementation as compared to other algorithms. [ABSTRACT FROM AUTHOR]

Discriminative feature extraction is a challenge for data-driven fault diagnosis. Although deep learning algorithms can automatically learn a good set of features without manual intervention, the lack of domain knowledge greatly limits the performance improvement, especially for nonstationary and nonlinear signals. This paper develops a multiscale information fusion-based stacked sparse autoencoder fault diagnosis method. The autoencoder takes advantage of the multiscale normalized frequency spectrum information obtained by dual-tree complex wavelet transform as input. Accordingly, the multiscale normalized features guarantee the translational invariance for signal characteristics, and the stacked sparse autoencoder benefits the unsupervised feature learning and ensures accurate and stable diagnosis performance. The developed method is performed on motor bearing vibration signals and worm gearbox vibration signals, respectively. The results confirm that the developed method can accommodate changing working conditions, be free of manual feature extraction, and perform better than the existing intelligent diagnosis methods. [ABSTRACT FROM AUTHOR]

The Industry 4.0 revolution is insisting strongly for use of machine learning-based processes and condition monitoring. In this paper, emphasis is given on machine learning-based approach for condition monitoring of shaft misalignment. This work highlights combined approach of artificial neural network and support vector machine for identification and measure of shaft misalignment. The measure of misalignment requires more features to be extracted under variable load conditions. Hence, primary objective is to measure misalignment with a minimum number of extracted features. This is achieved through normalization of vibration signal. An experimental setup is prepared to collect the required vibration signals. The normalized time domain nonstationary signals are given to discrete wavelet transform for features extraction. The extracted features such as detailed coefficient is considered for feature selection viz. Skewness, Kurtosis, Max, Min, Root mean square, and Entropy. The ReliefF algorithm is used to decide best feature on rank basis. The ratio of maximum energy to Shannon entropy is used in wavelet selection. The best feature is used to train machine learning algorithm. The rank-based feature selection has improved classification accuracy of support vector machine. The result obtained with the combined approach are discussed for different misalignment conditions. [ABSTRACT FROM AUTHOR]

Automatic modulation recognition has successfully used various machine learning methods and achieved certain results. As a subarea of machine learning, deep learning has made great progress in recent years and has made remarkable progress in the field of image and language processing. Deep learning requires a large amount of data support. As a communication field with a large amount of data, there is an inherent advantage of applying deep learning. However, the extensive application of deep learning in the field of communication has not yet been fully developed, especially in underwater acoustic communication. In this paper, we mainly discuss the modulation recognition process which is an important part of communication process by using the deep learning method. Different from the common machine learning methods that require feature extraction, the deep learning method does not require feature extraction and obtains more effects than common machine learning. [ABSTRACT FROM AUTHOR]

Summary: The classification of single and simultaneous power quality disturbances (PQDs) has become an issue of concern in the power system field. This paper proposes a novel approach based on dual strong tracking filters (STFs) and the rule‐based extreme learning machine (ELM) for detecting and classifying single and simultaneous PQDs. Dual STFs are a hybrid structure of a low‐order STF and high‐order STF. The fading factor of the low‐order STF is used to detect sudden changes in PQDs; the fundamental amplitude variation is tracked by the high‐order STF. Six distinctive features extracted from the dual STFs serve as the input to the ELM classifier for PQD classification. The rule‐based ELM technique, which is equipped with certain decision rules, can improve the ELM classification accuracy when the number of hidden nodes is insufficient. In consideration of special structures of matrices, the real‐time computation of the proposed method can be realized. A PQD dataset is generated in MATLAB for simulation experiments; the results show that 20 types of PQDs, including single and simultaneous disturbances, can be accurately classified under the different levels of noise via the proposed method. The method is also tested on a real recorded waveform to verify its effectiveness in PQD classification. [ABSTRACT FROM AUTHOR]

Multispectral palmprint recognition system (MPRS) is an essential technology for effective human identification and verification tasks. To improve the accuracy and performance of MPRS, a novel approach based on autoencoder (AE) and regularized extreme learning machine (RELM) is proposed in this paper. The proposed approach is intended to make the recognition faster by reducing the number of palmprint features without degrading the accuracy of classifier. To achieve this objective, first, the region of interest (ROI) from palmprint images is extracted by David Zhang’s method. Second, an efficient normalized Gist (NGist) descriptor is used for palmprint feature extraction. Then, the dimensionality of extracted features is reduced using optimized AE. Finally, the reduced features are fed to the RELM for classification. A comprehensive set of experiments are conducted on the benchmark MS-PolyU dataset. The results were significantly high compared to the state-of-the-art approaches, and the robustness and efficiency of the proposed approach are revealed. [ABSTRACT FROM AUTHOR]

Parkinson’s disease affects millions of people around the world and consequently various approaches have emerged to help diagnose this disease, among which we can highlight handwriting exams. Extracting features from handwriting exams is an important contribution of the computational field for the diagnosis of this disease. In this paper, we propose an approach that measures the similarity between the exam template and the handwritten trace of the patient following the exam template. This similarity was measured using the Structural Cooccurrence Matrix to calculate how close the handwritten trace of the patient is to the exam template. The proposed approach was evaluated using various exam templates and the handwritten traces of the patient. Each of these variations was used together with the Naïve Bayes, OPF, and SVM classifiers. In conclusion the proposed approach was proven to be better than the existing methods found in the literature and is therefore a promising tool for the diagnosis of Parkinson’s disease. [ABSTRACT FROM AUTHOR]

Preterm birth (PTB) is the leading cause of perinatal mortality and long-term morbidity, which results in significant health and economic problems. The early detection of PTB has great significance for its prevention. The electrohysterogram (EHG) related to uterine contraction is a noninvasive, real-time, and automatic novel technology which can be used to detect, diagnose, or predict PTB. This paper presents a method for feature extraction and classification of EHG between pregnancy and labour group, based on Hilbert-Huang transform (HHT) and extreme learning machine (ELM). For each sample, each channel was decomposed into a set of intrinsic mode functions (IMFs) using empirical mode decomposition (EMD). Then, the Hilbert transform was applied to IMF to obtain analytic function. The maximum amplitude of analytic function was extracted as feature. The identification model was constructed based on ELM. Experimental results reveal that the best classification performance of the proposed method can reach an accuracy of 88.00%, a sensitivity of 91.30%, and a specificity of 85.19%. The area under receiver operating characteristic (ROC) curve is 0.88. Finally, experimental results indicate that the method developed in this work could be effective in the classification of EHG between pregnancy and labour group. [ABSTRACT FROM AUTHOR]

A new optimized extreme learning machine- (ELM-) based method for power system transient stability prediction (TSP) using synchrophasors is presented in this paper. First, the input features symbolizing the transient stability of power systems are extracted from synchronized measurements. Then, an ELM classifier is employed to build the TSP model. And finally, the optimal parameters of the model are optimized by using the improved particle swarm optimization (IPSO) algorithm. The novelty of the proposal is in the fact that it improves the prediction performance of the ELM-based TSP model by using IPSO to optimize the parameters of the model with synchrophasors. And finally, based on the test results on both IEEE 39-bus system and a large-scale real power system, the correctness and validity of the presented approach are verified. [ABSTRACT FROM AUTHOR]