Search

This study presents findings indicating that the ferroelectric tunnel junction (FTJ) or resistive random-access memory (RRAM) in one cell can be intentionally selected depending on the application. The HfAlO film annealed at 700 °C shows stable FTJ characteristics and can be converted into RRAM by forming a conductive filament inside the same cell, that is, the process of intentionally forming a conductive filament is the result of defect generation and redistribution, and applying compliance current prior to a hard breakdown event of the dielectric film enables subsequent RRAM operation. The converted RRAM demonstrated good memory performance. Through current–voltage fitting, it was confirmed that the two resistance states of the FTJ and RRAM had different transport mechanisms. In the RRAM, the 1/f noise power of the high-resistance state (HRS) was about ten times higher than that of the low-resistance state (LRS). This is because the noise components increase due to the additional current paths in the HRS. The 1/f noise power according to resistance states in the FTJ was exactly the opposite result from the case of the RRAM. This is because the noise component due to the Poole–Frenkel emission is added to the noise component due to the tunneling current in the LRS. In addition, we confirmed the potentiation and depression characteristics of the two devices and further evaluated the accuracy of pattern recognition through a simulation by considering a dataset from the Modified National Institute of Standards and Technology. [ABSTRACT FROM AUTHOR]

We investigate a synaptic device with short-term memory characteristics using IGZO/SnOx as the switching layer. The thickness and components of each layer are analyzed by using x-ray photoelectron spectroscopy and transmission electron microscopy. The memristor exhibits analog resistive switching and a volatile feature with current decay over time. Moreover, through ten cycles of potentiation and depression, we demonstrate stable conductance modulation, leading to high-accuracy Modified National Institute of Standards and Technology pattern recognition. We effectively emulate the learning system of a biological synapse, including paired-pulse facilitation, spiking-amplitude-dependent plasticity, and spiking-rate-dependent plasticity (SRDP) by pulse trains. Ultimately, 4-bit reservoir computing divided into 16 states is incarnated using a pulse stream considering short-term memory plasticity and decay properties. [ABSTRACT FROM AUTHOR]

This study presents a preliminary exploration of thermally oxidized TaOx-based memristors and their potential as artificial synapses. Unlike the 10-min annealed devices, which display instability due to current overshoots, the 5-min annealed device exhibits stable resistive switching, retention, and endurance characteristics. Moreover, our memristor showcases synaptic behaviors encompassing potentiation, depression, spike-timing-dependent plasticity, and excitatory postsynaptic currents. This synaptic emulation holds tremendous promise for applications in neuromorphic computing, offering the opportunity to replicate the adaptive learning principles observed in biological synapses. In addition, we evaluate the device's suitability for pattern recognition within a neural network using the modified National Institute of Standards and Technology dataset. Our assessment reveals that the Pt/TaOx/Ta memristor with an oxidized insulator achieves outstanding potential manifested by an accuracy of 93.25% for the identical pulse scheme and an impressive accuracy of 95.42% for the incremental pulse scheme. [ABSTRACT FROM AUTHOR]

Introduction: The meta-learning methods have been widely used to solve the problem of few-shot learning. Generally, meta-learners are trained on a variety of tasks and then generalized to novel tasks. Methods: However, existing meta-learning methods do not consider the relationship between meta-tasks and novel tasks during the meta-training period, so that initial models of the meta-learner provide less useful meta- knowledge for the novel tasks. This leads to a weak generalization ability on novel tasks. Meanwhile, different initial models contain different meta-knowledge, which leads to certain differences in the learning effect of novel tasks during the-meta-testing period. Therefore, this article puts forward a meta-optimization method based on situational meta-task construction and cooperation of multiple initial models. First, during the meta-training period, a method of constructing situational meta-task is proposed, and the selected candidate task sets provide more effective meta-knowledge for novel tasks. Then, during the meta-testing period, an ensemble model method based on meta-optimization is proposed to minimize the loss of inter-model cooperation in prediction, so that multiple models cooperation can realize the learning of novel tasks. Results: The above-mentioned methods are applied to popular few-shot character datasets and image recognition datasets. Furthermore, the experiment results indicate that the proposed method achieves good effects in few-shot classification tasks. Discussion: In future work, we will extend our methods to provide more generalized and useful meta-knowledge to the model during the meta-training period when the novel few-shot tasks are completely invisible. [ABSTRACT FROM AUTHOR]

This article explores the use of convolutional neural networks (CNN) in studying retinal neural networks. It discusses the application of CNN as a framework for analyzing retinal circuits and examines the relationship between structural parameters and information-processing functions. The article focuses on the primary rod signal pathway in mouse and macaque retinas, particularly electrical synapses. By comparing the differences in retinal circuitry between these species, the author suggests that insights into neurocircuitry principles and adaptational designs can be gained. The article also discusses the methodology for observing neural circuitry using electron microscopy and emphasizes the importance of human pattern recognition in interpreting the images. It concludes by stating that connectomic studies on visual processing can enhance our understanding of natural intelligence. The document includes ethical guidelines, financial support, acknowledgments, a conflict of interest statement, and references to related research papers. [Extracted from the article]

The article presents the concepts of control systems that use methods of automatic analysis and interpretation of signals. These issues are presented through the description of three control strategies: simple closed-loop control, control using a process of classification, and using a signal understanding technique. The systems based on the discussed concepts are illustrated with examples of controlling a milling machine and an autonomous vehicle. In addition, in more detail, a task of learning human motor activities is described. This task, due to the nature of the controlled object, which is a human being, is extremely difficult. The article shows that the advanced control process, in which the control algorithm is selected and its parameters are adapted to the current situation, may be implemented through the use of the classification process and machine learning methods in general. Changing the algorithm is also possible using signal understanding techniques. These techniques, utilizing models of the objects, allow to predict the long-term effects of the control process. The ability to build control systems that operate in the above manner is of huge practical importance. The aim of this article is to describe the methods of automatic signal interpretation used in control processes and identify the main problems related to their use. The key problems refer to the acquisition of expert knowledge by the system. In order for this knowledge to be effectively transferred, the methods used in the system should have a high level of explainability. Showing the essential nature of this feature is the main outcome of this work. [ABSTRACT FROM AUTHOR]

Cloud Computing (CC) is a solution that efficiently maps cloud tasks to resources, ensuring high-performance utilization in computing. But, the challenge lies in managing attacks while managing a large data flow. For this reason, Lightweight deep learning model to secure authentication (LDLSA) was developed using Deep Convolutional Neural Network (DCNN) and Homomorphic Encryption (HE) is developed for deep face recognition authentication, enhancing privacy protection in CC. However, these models are computationally expensive, require extensive training data and lack information on cloud-stored authentication data for privacy preservation. In this paper, DL with Cloud Authentication (dCAuth) is proposed to resolve the above mentioned issues to provide efficient cloud authentication with lesser complexity. This method employs a spurring algorithm for dwindling DCNN with recall (dDCNNr) to handle the number of parameters in DCNN to handle a large number of parameters for cloud authentication. dDCNNr model reduces execution time and improves classification accuracy, but it necessitates faster training time. To solve this, Fully Connected (FC) layers of DCNN is replaced with Hopfield Neural Network (HNN). The model utilises a DCNN for feature extraction and HNN for pattern recognition highly focusing on authentication, both the process have similar training phase with known patterns by adjusted weights. The proposed authentication method stores encrypted Neural Network (NN) weights in the cloud, eliminating the need for a verification table. The weights for unknown input patterns are generated during pattern recall. The HNN weight matches input indicating a known or legal pattern, aiming to identify a known pattern that best fits the input with minimum processing resources. The proposed authentication model integrates DCNN and HNN to rapidly and precisely recall the legitimate user ID and face image (password) information. The model effectively manages the large training data and parameters, ensuring privacy for cloud-stored authentication data, simultaneously reducing the time duration of registration and password changes (old to new face image). Finally, an extensive simulation reveals that the proposed model achieves accuracy of 94.15%, 94.53% and 94.38% on Georgia Tech face (GTF), Labelled Faces in the Wild (LFW) and Biometric Signature (BS) database respectively. [ABSTRACT FROM AUTHOR]

The learning speed of feed-forward neural networks is notoriously slow and has presented a bottleneck in deep learning applications for several decades. For instance, gradient-based learning algorithms, which are used extensively to train neural networks, tend to work slowly when all of the network parameters must be iteratively tuned. To counter this, both researchers and practitioners have tried introducing randomness to reduce the learning requirement. Based on the original construction of Igelnik and Pao, single layer neural-networks with random input-to-hidden layer weights and biases have seen success in practice, but the necessary theoretical justification is lacking. In this study, we begin to fill this theoretical gap. We then extend this result to the non-asymptotic setting using a concentration inequality for Monte-Carlo integral approximations. We provide a (corrected) rigorous proof that the Igelnik and Pao construction is a universal approximator for continuous functions on compact domains, with approximation error squared decaying asymptotically like O(1/n) for the number n of network nodes. We then extend this result to the non-asymptotic setting, proving that one can achieve any desired approximation error with high probability provided n is sufficiently large. We further adapt this randomized neural network architecture to approximate functions on smooth, compact submanifolds of Euclidean space, providing theoretical guarantees in both the asymptotic and non-asymptotic forms. Finally, we illustrate our results on manifolds with numerical experiments. [ABSTRACT FROM AUTHOR]

Existing segmentation‐based methods have made considerable progress in arbitrarily shaped text detection due to the advantage of dealing with shape variation. However, there still exist challenges to detecting accurate text instances with dense layouts, inaccurate annotations, and complex backgrounds. Many recent works have focused on improving arbitrary boundary prediction, but it may be difficult to accurately distinguish each instance of dense layouts because their boundary pixels may be mistakenly classified to produce inaccurate results (i.e., adhesive texts) with inaccurate annotation and complex backgrounds. Considering the local and long‐range dependencies, this paper proposes an efficient text detector, namely T‐Skeleton, to obtain more reliable segmentation detections. In the spirit of object skeletonization, we introduce the text instance skeleton highlighting the semantically significant structure (similar to the skeleton of a fish) to explicitly capture the long‐range dependencies of text instances. The key idea of T‐Skeleton is to calibrate the coarse text proposals by embedding text instance skeletons to separate crowd texts accurately and robustly. We further design a channel attention module to enlarge the performance margin between T‐Skeleton and the segmentation baseline. Experimental results on four publicly available datasets show the superiority of T‐Skeleton in handling long and curved texts. [ABSTRACT FROM AUTHOR]

The Neutrosophic Numeral System Algorithms are a set of techniques designed to handle uncertainty and ambiguity in numerical data. These algorithms use Neutrosophic Set Theory, a mathematical framework that deals with incomplete, indeterminate, and inconsistent information. In this paper, we provide an overview of different approaches used in Neutrosophic Numeral System Algorithms, including Neutrosophic Binary System, Neutrosophic Decimal System, and Neutrosophic Octal System. These systems use different bases and representations to account for degrees of truth, indeterminacy, and falsity in numerical data. We also explore the relationship between Neutrosophic Numeral System Algorithms and Number Neutrosophic Systems, which are another type of Neutrosophic System used for representing numerical data. Number Neutrosophic Systems use Neutrosophic Numbers to represent degrees of truth, indeterminacy, and falsity in numerical data, and they can be used in conjunction with Neutrosophic Numeral System Algorithms to handle uncertainty and ambiguity in decision-making and artificial intelligence applications. Moreover. We discuss the advantages and disadvantages of each algorithm and their potential applications in various fields. Finally, we highlight the importance of Neutrosophic cryptography in addressing uncertainty and ambiguity in decision making and artificial intelligence and discuss future research directions. Understanding Neutrosophic Numeral System Algorithms and their relationship with Number Neutrosophic Systems is crucial for developing effective techniques for handling uncertainty and ambiguity in numerical data in decision-making, pattern recognition, and artificial intelligence applications. [ABSTRACT FROM AUTHOR]

In complex network analyses, mainly in social networks, the detection of communities is an important source of information for revealing an internal organization of nodes. On the other hand, if the network reveals a leadership structure, it is possible to understand the mechanisms of information dissemination on it. The detection of leaders and communities is a big challenge depending on the complexity level of the network. In the literature, there are some metaheuristic algorithms for detecting leaders and communities based on pattern recognition on the graph associated with the network. In this paper, we developed a random graph system to generate a synthetic network instance with leader and community structures that define a ground truth. We compare this ground truth to determine the performance of the algorithms LCDA 1 and LCDA 2 for detecting leaders and communities. The results corroborate that the benchmarking system would help in selecting useful configurations for practical applications. [ABSTRACT FROM AUTHOR]

Education is crucial for a productive life and providing necessary resources. With the advent of technology like artificial intelligence, higher education institutions are incorporating technology into traditional teaching methods. Predicting academic success has gained interest in education as a strong academic record improves a university's ranking and increases student employment opportunities. Modern learning institutions face challenges in analyzing performance, providing high-quality education, formulating strategies for evaluating students' performance, and identifying future needs. E-learning is a rapidly growing and advanced form of education, where students enroll in online courses. Platforms like Intelligent Tutoring Systems (ITS), learning management systems (LMS), and massive open online courses (MOOC) use educational data mining (EDM) to develop automatic grading systems, recommenders, and adaptative systems. However, e-learning is still considered a challenging learning environment due to the lack of direct interaction between students and course instructors. Machine learning (ML) is used in developing adaptive intelligent systems that can perform complex tasks beyond human abilities. Some areas of applications of ML algorithms include cluster analysis, pattern recognition, image processing, natural language processing, and medical diagnostics. In this research work, K-means, a clustering data mining technique using Davies' Bouldin method, obtains clusters to find important features affecting students' performance. The study found that the SVM algorithm had the best prediction results after parameter adjustment, with a 96% accuracy rate. In this paper, the researchers have examined the functions of the Support Vector Machine, Decision Tree, naive Bayes, and KNN classifiers. The outcomes of parameter adjustment greatly increased the accuracy of the four prediction models. Naïve Bayes model's prediction accuracy is the lowest when compared to other prediction methods, as it assumes a strong independent relationship between features. [ABSTRACT FROM AUTHOR]

The construction of modern power system is key to achieving dual carbon goals, where non-intrusive load monitoring (NILM) plays a vital role in enhancing energy utilization efficiency and energy management. For example, to enable prosumers to better understand the extent of their flexible loads for demand response and peer-to-peer trading, it is essential to be aware of the types and states of loads using the method of NILM. To improve the predictive accuracy and implementation effectiveness of NILM technology, this paper proposes a novel NILM method integrating meteorological and calendar features. It delves deeply into the close connection between external factors such as temperature, precipitation, wind speed, and holidays, and the energy consumption of electrical appliances, constructing additional associative mappings in the training of the Denoising Autoencoder (DAE) model. Test results on the UK-DALE public dataset show that the NILM method proposed in this paper has significant advantages over traditional NILM methods that consider only single-dimensional electrical data features, in terms of load pattern recognition and accuracy in load energy consumption monitoring. This confirms the potential of multi-dimensional feature fusion technology in the application of NILM. [ABSTRACT FROM AUTHOR]

Handwritten character recognition is one of the classical problems in the field of image classification. Supervised learning techniques using deep learning models are highly effective in their application to handwritten character recognition. However, they require a large dataset of labeled samples to achieve good accuracies. Recent supervised learning techniques for Kannada handwritten character recognition have state of the art accuracy and perform well over a large range of input variations. In this work, a framework is proposed for the Kannada language that incorporates techniques from semi-supervised learning. The framework uses features extracted from a convolutional neural network backbone and uses regularization to improve the trained features and label propagation to classify previously unseen characters. The episodic learning framework is used to validate the framework. Twenty-four classes are used for pre-training, 12 classes are used for testing and 11 classes are used for validation. Fine-tuning is tested using one example per unseen class and five examples per unseen class. Through experimentation the components of the network are implemented in Python using the Pytorch library. It is shown that the accuracy obtained 99.13% make this framework competitive with the currently available supervised learning counterparts, despite the large reduction in the number of labeled samples available for the novel classes. [ABSTRACT FROM AUTHOR]

This editorial published in the journal Frontiers in Psychology discusses the importance of early indicators of cognitive decline, Alzheimer's disease, and related dementias. It emphasizes the need for early detection and intervention, particularly in the mild cognitive impairment stage. The editorial presents insights from various research projects and reviews that focus on neurophysiological tools such as EEG, eye-tracking, and optical coherence tomography angiography. These tools have the potential to provide valuable insights into the functioning of the brain and potential indicators of neurodegenerative conditions. The authors express their gratitude to the reviewers, editorial board, and publication specialists for their support in enhancing the quality of this research. [Extracted from the article]

In this study, the development and characterization of 2D ferroelectric field‐effect transistor (2D FeFET) devices are presented, utilizing nanoscale ferroelectric HfZrO2 (HZO) and 2D semiconductors. The fabricated device demonstrated multi‐level data storage capabilities. It successfully emulated essential biological characteristics, including excitatory/inhibitory postsynaptic currents (EPSC/IPSC), Pair‐Pulse Facilitation (PPF), and Spike‐Timing Dependent Plasticity (STDP). Extensive endurance tests ensured robust stability (107 switching cycles, 105 s (extrapolated to 10 years)), excellent linearity, and high Gmax/Gmin ratio (>105), all of which are essential for realizing multi‐level data states (>7‐bit operation). Beyond mimicking synaptic functionalities, the device achieved a pattern recognition accuracy of ≈94% on the Modified National Institute of Standards and Technology (MNIST) handwritten dataset when incorporated into a neural network, demonstrating its potential as an effective component in neuromorphic systems. The successful implementation of the 2D FeFET device paves the way for the development of high‐efficiency, ultralow‐power neuromorphic hardware which is in sub‐femtojoule (48 aJ/spike) and fast response (1 µs), which is 104 folds faster than human synapse (≈10 ms). The results of the research underline the potential of nanoscale ferroelectric and 2D materials in building the next generation of artificial intelligence technologies. [ABSTRACT FROM AUTHOR]

This paper presents a novel webcam-based approach for gaze estimation on computer screens. Utilizing appearance based gaze estimation models, the system provides a method for mapping the gaze vector from the user's perspective onto the computer screen. Notably, it determines the user's 3D position in front of the screen, using only a 2D webcam without the need for additional markers or equipment. The study presents a comprehensive comparative analysis, assessing the performance of the proposed method against established eye tracking solutions. This includes a direct comparison with the purpose-built Tobii Eye Tracker 5, a high-end hardware solution, and the webcam-based GazeRecorder software. In experiments replicating head movements, especially those imitating yaw rotations, the study brings to light the inherent difficulties associated with tracking such motions using 2D webcams. This research introduces a solution by integrating Structure from Motion (SfM) into the Convolutional Neural Network (CNN) model. The study's accomplishments include showcasing the potential for accurate screen gaze tracking with a simple webcam, presenting a novel approach for physical distance computation, and proposing compensation for head movements, laying the groundwork for advancements in real-world gaze estimation scenarios. [ABSTRACT FROM AUTHOR]

Accurate medical recordkeeping is important for personal and public health. Conventional forms of on‐patient medical information, such as medical alert bracelets or finger‐markings, may compromise patient privacy because they are readily visible to other people. Here, the development of an invisible, temporary, and easily deployable on‐patient medical recordkeeping system is reported. Information is stored in unique patterns of spatially distributed near‐infrared (NIR) fluorescent quantum dots (QDs), which are delivered to the skin using dissolvable microneedle arrays. The patterns are invisible to the naked eye but detectable with an infrared camera, which can extract information with >98% accuracy using automated pattern recognition software. By encapsulating NIR QDs in an FDA‐approved biodegradable polymer, biodegradation rates can be tuned so that the encoded medical information can be conveyed in both a spatial and temporal manner, with some components fading within 100 days and others persisting for 6 months. This may be particularly useful for administering a series of vaccinations or treatments by indicating if enough time has passed for the patient to receive the next dose. Importantly, this system contains no personal information, does not require connection to a centralized database, and is not visible to the naked eye, ensuring patient privacy. [ABSTRACT FROM AUTHOR]

Pedestrian Dead Reckoning (PDR) is a promising algorithm for indoor positioning. However, the accuracy of PDR degrades due to the accumulated error, especially in multi-floor buildings. This paper introduces a three-dimensional (3D) positioning method based on terrain feature matching to reduce the influence of accumulated errors in multi-floor scenes. The proposed calibration method involves two steps: motion pattern recognition and position matching-based calibration. The motion pattern recognition aims to detect different motion patterns, i.e., taking the stairs or horizontal walking, from the streaming data. Then, stair entrances and corridor corners are matched with transition points of motion patterns and pedestrian turning points, respectively. After matching, calibration is performed to eliminate the accumulated errors. By carrying out experiments on a two-floor closed-loop path with a walking distance about 145 m, it is shown that this method can effectively reduce the accumulated error of PDR, achieving accurate 3D positioning. The average error is reduced from 6.60 m to 1.37 m. [ABSTRACT FROM AUTHOR]

Non-intrusive load monitoring (NILM) has emerged as a pivotal technology in energy management applications by enabling precise monitoring of individual appliance energy consumption without the requirements of intrusive sensors or smart meters. In this technique, the load disaggregation for the individual device is accrued by the recognition of their current signals by employing machine learning (ML) methods. This research paper conducts a comprehensive comparative analysis of various ML techniques applied to NILM, aiming to identify the most effective methodologies for accurate load disaggregation. The study employs a diverse dataset comprising high-resolution electricity consumption data collected from an Estonian household. The ML algorithms, including deep neural networks based on long short-term memory networks (LSTM), extreme gradient boost (XgBoost), logistic regression (LR), and dynamic time warping with K-nearest neighbor (DTW-KNN) are implemented and evaluated for their performance in load disaggregation. Key evaluation metrics such as accuracy, precision, recall, and F1 score are utilized to assess the effectiveness of each technique in capturing the nuanced energy consumption patterns of diverse appliances. Results indicate that the XgBoost-based model demonstrates superior performance in accurately identifying and disaggregating individual loads from aggregated energy consumption data. Insights derived from this research contribute to the optimization of NILM techniques for real-world applications, facilitating enhanced energy efficiency and informed decision-making in smart grid environments. [ABSTRACT FROM AUTHOR]

Partial discharge detection is considered a crucial technique for evaluating insulation performance and identifying defect types in cable terminals of high-speed electric multiple units (EMUs). In this study, terminal samples exhibiting four typical defects were prepared from high-speed EMUs. A cable discharge testing system, utilizing high-frequency current sensing, was developed to collect discharge signals, and datasets corresponding to these defects were established. This study proposes the use of the convolutional neural network (CNN) for the classification of discharge signals associated with specific defects, comparing this method with two existing neural network (NN)-based classification models that employ the back-propagation NN and the radial basis function NN, respectively. The comparative results demonstrate that the CNN-based model excels in accurately identifying signals from various defect types in the cable terminals of high-speed EMUs, surpassing the two existing NN-based classification models. [ABSTRACT FROM AUTHOR]

This document is a summary of a special issue of the journal Sensors titled "Artificial Intelligence Methods for Smart Cities." The special issue focuses on the use of artificial intelligence (AI) and the Internet of Things (IoT) in smart cities. The articles in the special issue cover a range of topics, including anomaly detection, traffic management, public safety, surveillance, and indoor surveillance. The articles highlight the potential benefits of AI and IoT in improving urban systems and addressing key urban challenges. The special issue also acknowledges the challenges and ethical considerations associated with the widespread adoption of these technologies. [Extracted from the article]

In the task of upper-limb pattern recognition, effective feature extraction, channel selection, and classification methods are crucial for the construction of an efficient surface electromyography (sEMG) signal classification framework. However, existing deep learning models often face limitations due to improper channel selection methods and overly specific designs, leading to high computational complexity and limited scalability. To address this challenge, this study introduces a deep learning network based on channel feature compression—partial channel selection sEMG net (PCS-EMGNet). This network combines channel feature compression (channel selection) and feature extraction (partial block), aiming to reduce the model's parameter count while maintaining recognition accuracy. PCS-EMGNet extracts high-dimensional feature vectors from sEMG signals through the partial block, decoding spatial and temporal feature information. Subsequently, channel selection compresses and filters these high-dimensional feature vectors, accurately selecting channel features to reduce the model's parameter count, thereby decreasing computational complexity and enhancing the model's processing speed. Moreover, the proposed method ensures the stability of classification, further improving the model's capability of recognizing features in sEMG signal data. Experimental validation was conducted on five benchmark databases, namely the NinaPro DB4, NinaPro DB5, BioPatRec DB1, BioPatRec DB2, and BioPatRec DB3 datasets. Compared to traditional gesture recognition methods, PCS-EMGNet significantly enhanced recognition accuracy and computational efficiency, broadening its application prospects in real-world settings. The experimental results showed that our model achieved the highest average accuracy of 88.34% across these databases, marking a 9.96% increase in average accuracy compared to models with similar parameter counts. Simultaneously, our model's parameter size was reduced by an average of 80% compared to previous gesture recognition models, demonstrating the effectiveness of channel feature compression in maintaining recognition accuracy while significantly reducing the parameter count. [ABSTRACT FROM AUTHOR]

Featured Application: This work is being developed as part of a digital health system designed to assist in the therapeutic treatment of people with autism spectrum disorder. Digital health apps have become a staple in daily life, promoting awareness and providing motivation for a healthier lifestyle. With an already overwhelmed healthcare system, digital therapies offer relief to both patient and physician alike. One such planned digital therapy application is the incorporation of an emotion recognition model as a tool for therapeutic interventions for people with autism spectrum disorder (ASD). Diagnoses of ASD have increased relatively rapidly in recent years. To ensure effective recognition of expressions, a system is designed to analyze and classify different emotions from facial landmarks. Facial landmarks combined with a corresponding mesh have the potential of bypassing hurdles of model robustness commonly affecting emotion recognition from images. Landmarks are extracted from facial images using the Mediapipe framework, after which a custom mesh is constructed from the detected landmarks and used as input to a graph convolution network (GCN) model for emotion classification. The GCN makes use of the relations formed from the mesh along with the special distance features extracted. A weighted loss approach is also utilized to reduce the effects of an imbalanced dataset. The model was trained and evaluated with the Aff-Wild2 database. The results yielded a 58.76% mean accuracy on the selected validation set. The proposed approach shows the potential and limitations of using GCNs for emotion recognition in real-world scenarios. [ABSTRACT FROM AUTHOR]

This article discusses the use of artificial intelligence (AI) in cardiovascular imaging and provides a framework for assessing the quality and clinical value of AI applications in this field. The article explains the definitions of AI, machine learning (ML), and deep learning (DL), and discusses their roles in cardiology. It emphasizes the importance of high-quality input data, addressing bias, and performing appropriate external validation. The article also highlights the need for interpretability and explainability in AI algorithms, the importance of addressing bias and promoting equity, and the evolving scientific, ethical, and regulatory landscape surrounding AI in healthcare. The authors suggest that a synergistic deployment of AI tools and trained professionals can benefit patient care, but caution against the misuse of AI. [Extracted from the article]

Well test model identification is a challenging task due to the numerous types of well test interpretation models and the non-uniqueness of pressure responses generated by different reservoir models. An automated framework is crucial to aid in the identification of well test interpretation models. Since the identification of well test interpretation relies primarily on the various flow regimes appeared on different diagnostic plots. A novel approach is proposed for the well test model identification from the pressure transient test data using the syntactic pattern recognition in this study. In this study, the identification process of well test interpretation model is divided into six steps: preprocessing, feature primitive extraction, curve shape tracking, flow regime division, model preliminary inference, and model final validation incorporating TDS technology. The automatic identification framework developed with this method has been able to identify a variety of complex well test interpretation models correctly, and the non-uniqueness of model results can be well resolved by syntactic pattern recognition combined with TDS technology. In general, the findings of this study can help for better understanding of the process by which well test expert completes the task of model identification. [ABSTRACT FROM AUTHOR]

Night pedestrian detection with visible image only suffers from the dilemma of high miss rate due to poor illumination conditions. Cross-modality fusion can ameliorate this dilemma by providing complementary information to each other through infrared and visible images. In this paper, we propose a cross-modal fusion framework based on YOLOv5, which is aimed at addressing the challenges of night pedestrian detection under low-light conditions. The framework employs a dual-stream architecture that processes visible images and infrared images separately. Through the Cross-Modal Feature Rectification Module (CMFRM), visible and infrared features are finely tuned on a granular level, leveraging their spatial correlations to focus on complementary information and substantially reduce uncertainty and noise from different modalities. Additionally, we have introduced a two-stage Feature Fusion Module (FFM), with the first stage introducing a cross-attention mechanism for cross-modal global reasoning, and the second stage using a mixed channel embedding to produce enhanced feature outputs. Moreover, our method involves multidimensional interaction, not only correcting feature maps in terms of channel and spatial dimensions but also applying cross-attention at the sequence processing level, which is critical for the effective generalization of cross-modal feature combinations. In summary, our research significantly enhances the accuracy and robustness of nighttime pedestrian detection, offering new perspectives and technical pathways for visual information processing in lowlight environments. [ABSTRACT FROM AUTHOR]

At present, the sorting work of library books heavily relies on manual labour, and the sorting process is time-consuming and laborious. To address the above issues, in the context of digital transformation in the information age, a strategy research of intelligent library service transformation based on convolutional network is proposed to achieve intelligent daily maintenance regarding libraries. In particular, the arrangement of books based on category and order is resolved. First, the Canny operator detection algorithm combined with straight line is used to extract the book spine from the book image. Second, the Faster-RCNN-based target detection algorithm is selected to locate the claim number. The positioning accuracy also reaches 96% in the positioning test, which is also better than other algorithms. The specific detection time is 100s. Then the method is used to segment the characters, and the accuracy is 97%. Finally, in the character recognition of library book request numbers, the feature extraction effect of character images based on convolutional networks is the best. Compared with 95% and 96% of other algorithms, the average recognition rate is 97%, which also improves network performance. The comparative experimental findings suggest that the convolutional network algorithm can effectively achieve the information detection and book order recognition of library common claim numbers, providing a feasible solution for library intelligence. [ABSTRACT FROM AUTHOR]

Due to the high cost of data acquisition in many specific fields, such as intelligent agriculture, the available data is insufficient for the typical deep learning paradigm to show its superior performance. As an important complement to deep learning, few-shot learning focuses on pattern recognition tasks under the constraint of limited data, which can be used to solve practical problems in many application fields with data scarcity. This survey summarizes the research status, main models and representative achievements of few-shot learning from four aspects: model fine-tuning, meta-learning, metric learning and data enhancement, and especially introduces the few-shot learning-driven typical applications in intelligent agriculture. Finally, the current challenges of few-shot learning and its development trends in intelligent agriculture are prospected. [ABSTRACT FROM AUTHOR]

Handwritten character recognition (HCR) involves identifying characters in images, documents, and various sources such as forms surveys, questionnaires, and signatures, and transforming them into a machine-readable format for subsequent processing. Successfully recognizing complex and intricately shaped handwritten characters remains a significant obstacle. The use of convolutional neural network (CNN) in recent developments has notably advanced HCR, leveraging the ability to extract discriminative features from extensive sets of raw data. Because of the absence of pre-existing datasets in the Kurdish language, we created a Kurdish handwritten dataset called (KurdSet). The dataset consists of Kurdish characters, digits, texts, and symbols. The dataset consists of 1560 participants and contains 45,240 characters. In this study, we chose characters only from our dataset. We utilized a Kurdish dataset for handwritten character recognition. The study also utilizes various models, including InceptionV3, Xception, DenseNet121, and a customCNNmodel. To show the performance of the KurdSet dataset, we compared it to Arabic handwritten character recognition dataset (AHCD). We applied the models to both datasets to show the performance of our dataset. Additionally, the performance of the models is evaluated using test accuracy, which measures the percentage of correctly classified characters in the evaluation phase. All models performed well in the training phase, DenseNet121 exhibited the highest accuracy among the models, achieving a high accuracy of 99.80% on the Kurdish dataset. And Xception model achieved 98.66% using the Arabic dataset. [ABSTRACT FROM AUTHOR]

The similarity measure of intuitionistic fuzzy sets is a primary method for dealing with uncertainty and fuzzy problems and is commonly used in fuzzy decision-making and pattern recognition. The current mainstream similarity measure is based on the classical fuzzy set with only one negation, which often violates the intuitionistic problem in applications because the actual semantics of multiple negations are not considered. To solve the inconsistency and irrationality problems in the classical similarity methods, we introduce three negations (contradiction negation, opposition negation, and mediation negation) in the intuitionistic fuzzy set to obtain the generalized intuitionistic fuzzy set and prove its related property theorem. On this basis, our similarity measure adopts a mediational negation to represent non-membership, which fully utilizes the multiple negation information of non-membership and hesitancy and avoids the loss of fuzzy information. We verify the method's rationality, validity, and originality through pattern recognition experiments and numerical examples, which improves the performance of intuitionistic fuzzy set similarity in practical applications and provides a new approach for future research on intuitionistic fuzzy inference. [ABSTRACT FROM AUTHOR]

Automated reading of license plate and its detection is a crucial component of the competent transportation system. Toll payment and parking management e-payment systems may benefit from this software's features. License plate detection and identification algorithms abound, and each has its own set of strengths and weaknesses. Computer vision has advanced rapidly in terms of new breakthroughs and techniques thanks to the emergence and proliferation of deep learning principles across several branches of AI. The practice of automating the monitoring process in traffic management, parking management, and police surveillance has become much more effective thanks to the development of Automatic License Plate Recognition (ALPR). Even though license plate recognition (LPR) is a technology that is extensively utilized and has been developed, there is still a significant amount of work to be done before it can achieve its full potential. In the last several years, there have been substantial advancements in both the scientific community's methodology and its level of efficiency. In this era of deep learning, there have been numerous developments and techniques established for LPR, and the purpose of this research is to review and examine those developments and approaches. In light of this, the authors of this study suggest a four-stage technique to automated license plate detection and identification (ALPDR), which includes, image pre-processing, license plate extraction, character segmentation, and character recognition. And the first three phases are known as "extraction," "pre-processing," and "segmentation," and each of these processes has been shown to benefit from its own unique technique. In light of the fact that character recognition is an essential component of license plate identification and detection, the Convolution Neural Network (CNN), MobileNet, Inception V3, and ResNet 50 have all been put through their paces in this regard. [ABSTRACT FROM AUTHOR]

A Single-valued neutrosophic set (SVNS) has recently been explored as a comprehensive tool to assess uncertain information due to varied human cognition. This notion stretches the domain of application of the classical fuzzy set and its extended versions. Various comparison measures based on SVNSs like distance measure, similarity measure, and, divergence measure have practical significance in the study of clustering analysis, pattern recognition, machine learning, and computer vision-related problems. Existing measures have some drawbacks in terms of precision and exclusion of information and produce unreasonable results in categorization problems. In this paper, we propose a generic method to define new divergence measures based on common aggregation operators and discuss some algebraic properties of the proposed divergence measures. To further appreciate the proposed divergence measures, their application to pattern recognition has been investigated in conjunction with the prominent existing comparison measures based on SVNSs. The comparative assessment sensitivity analysis of the proposed measures establishes their edge over the existing ones because of appropriate classification results. [ABSTRACT FROM AUTHOR]

Analyzing point clouds with neural networks is a current research hotspot. In order to analyze the 3D geometric features of point clouds, most neural networks improve the network performance by adding local geometric operators and trainable parameters. However, deep learning usually requires a large amount of computational resources for training and inference, which poses challenges to hardware devices and energy consumption. Therefore, some researches have started to try to use a nonparametric approach to extract features. Point-NN combines nonparametric modules to build a nonparametric network for 3D point cloud analysis, and the nonparametric components include operations such as trigonometric embedding, farthest point sampling (FPS), k-nearest neighbor (k-NN), and pooling. However, Point-NN has some blindness in feature embedding using the trigonometric function during feature extraction. To eliminate this blindness as much as possible, we utilize a nonparametric energy function-based attention mechanism (ResSimAM). The embedded features are enhanced by calculating the energy of the features by the energy function, and then the ResSimAM is used to enhance the weights of the embedded features by the energy to enhance the features without adding any parameters to the original network; Point-NN needs to compute the similarity between each feature at the naive feature similarity matching stage; however, the magnitude difference of the features in vector space during the feature extraction stage may affect the final matching result. We use the Squash operation to squeeze the features. This nonlinear operation can make the features squeeze to a certain range without changing the original direction in the vector space, thus eliminating the effect of feature magnitude, and we can ultimately better complete the naive feature matching in the vector space. We inserted these modules into the network and build a nonparametric network, Point-Sim, which performs well in 3D classification tasks. Based on this, we extend the lightweight neural network Point-SimP by adding some trainable parameters for the point cloud classification task, which requires only 0.8 M parameters for high performance analysis. Experimental results demonstrate the effectiveness of our proposed algorithm in the point cloud shape classification task. The corresponding results on ModelNet40 and ScanObjectNN are 83.9% and 66.3% for 0 M parameters—without any training—and 93.3% and 86.6% for 0.8 M parameters. The Point-SimP reaches a test speed of 962 samples per second on the ModelNet40 dataset. The experimental results show that our proposed method effectively improves the performance on point cloud classification networks. [ABSTRACT FROM AUTHOR]

Ground‐penetrating radar is an effective tool for detecting landmines and improvised explosive devices, but its performance is strongly influenced by subsurface properties as well as the characteristics of the target. To complement or replace labour‐intensive experiments on test sites, cost‐efficient electromagnetic wave propagation simulations using the finite‐difference time‐domain method are being increasingly used. However, to obtain realistic synthetic data, accurate modelling of signal alteration caused by dispersion, scattering from soil material, target contrast, shape, and inner setup, as well as the coupling effects of the antenna to the ground is required. In this study, we present a detailed three‐dimensional model of a shielded ground‐penetrating radar antenna applied to various scenarios containing metallic and non‐metallic targets buried in different soils. The frequency‐dependent intrinsic material properties of soil samples were measured with the coaxial transmission‐line technique, while a discrete random media was used to implement the heterogeneity of a gravel based on its grain‐size distribution. Our simulations show very good agreement with experimental validation data collected under controlled conditions. We accurately reproduce the amplitude, frequency, and phase of target signals, the subsurface background noise, antenna crosstalk and associated interference with target signals, and the effect of antenna elevation. The approach allows for a systematic investigation of the effects of soil, target, and sensor properties on detection performance, providing insight into novel and complex ground‐penetrating radar scenarios and the potential for a wide range of simulation possibilities for demining with ground‐penetrating radar. These investigations have the potential to improve the safety and effectiveness of landmine and improvised explosive device detection in the future, such as building a database for training deminers or developing automatic signal pattern recognition algorithms. [ABSTRACT FROM AUTHOR]

Heart disease (HD) accounts for more deaths every year than other illnesses. World Health Organization (WHO) assessed 17.9 million life losses caused by heart disease in 2016, demonstrating 31% of all international life losses. Three-quarters of these life losses occur in low and middle-income nations. Machine learning (ML), due to advanced precision in pattern recognition and classification, demonstrates to be in effect in complementing decision-making and threat prediction from the huge number of HD data created by the healthcare sector. Thus, this study aims to develop a logistic regression model (LRM) for predicting the risk of getting HD in ten years. The study explores the different methodologies for improving the performance of base LRM for predicting whether a person gets HD after ten years or not. The result demonstrates the capability of LRM in predicting the risks of getting HD after ten years. The LRM achieves 97.35% accuracy with the recursive feature elimination and random under-sampling. This implies that the LRM can play an important role in precautionary methods to avoid the risk of HD. [ABSTRACT FROM AUTHOR]

Current computing systems rely on Boolean logic and von Neumann architecture, where computing cells are based on high-speed electron-conducting complementary metal-oxide-semiconductor (CMOS) transistors. In contrast, ions play an essential role in biological neural computing. Compared with CMOS units, the synapse/neuron computing speed is much lower, but the human brain performs much better in many tasks such as pattern recognition and decision-making. Recently, ionic dynamics in oxide electrolyte-gated transistors have attracted increasing attention in the field of neuromorphic computing, which is more similar to the computing modality in the biological brain. In this review article, we start with the introduction of some ionic processes in biological brain computing. Then, electrolyte-gated ionic transistors, especially oxide ionic transistors, are briefly introduced. Later, we review the state-of-the-art progress in oxide electrolyte-gated transistors for ionic neuromorphic computing including dynamic synaptic plasticity emulation, spatiotemporal information processing, and artificial sensory neuron function implementation. Finally, we will address the current challenges and offer recommendations along with potential research directions. [ABSTRACT FROM AUTHOR]

In the context of escalating energy demands and the quest for sustainable waste management solutions, this paper evaluates the efficacy of three machine learning methods—ElasticNet, Decision Trees, and Neural Networks—in predicting energy recovery from municipal waste across the European Union. As renewable energy sources increasingly dominate the energy production landscape, the integration of Waste-to-Energy (WTE) processes presents a dual advantage: enhancing waste management and contributing to the renewable energy mix. This study leverages a dataset incorporating economic and environmental indicators from 25 European countries, spanning 2013–2020, to compare the predictive capabilities of the three machine learning models. The analysis reveals that Neural Networks, with their intricate pattern recognition capabilities, outperform ElasticNet and Decision Trees in predicting energy recovery metrics, as evidenced by superior performance in key statistical indicators such as R-value, Mean Squared Error (MSE), and Mean Absolute Error (MAE). The comparative analysis not only demonstrates the effectiveness of each method but also suggests Neural Networks as a pivotal tool for informed decision-making in waste management and energy policy formulation. Through this investigation, the paper contributes to the sustainable energy and waste management discourse, emphasizing the critical intersection of advanced technologies, policy considerations, and environmental stewardship in addressing contemporary energy challenges. [ABSTRACT FROM AUTHOR]

In this study, we delve into a novel approach by employing a sensor-based pattern recognition model to address the automation of construction equipment activity analysis. The model integrates time of flight (ToF) sensors with deep convolutional neural networks (DCNNs) to accurately classify the operational activities of construction equipment, focusing on piston movements. The research utilized a one-twelfth-scale excavator model, processing the displacement ratios of its pistons into a unified dataset for analysis. Methodologically, the study outlines the setup of the sensor modules and their integration with a controller, emphasizing the precision in capturing equipment dynamics. The DCNN model, characterized by its four-layered convolutional blocks, was meticulously tuned within the MATLAB environment, demonstrating the model's learning capabilities through hyperparameter optimization. An analysis of 2070 samples representing six distinct excavator activities yielded an impressive average precision of 95.51% and a recall of 95.31%, with an overall model accuracy of 95.19%. When compared against other vision-based and accelerometer-based methods, the proposed model showcases enhanced performance and reliability under controlled experimental conditions. This substantiates its potential for practical application in real-world construction scenarios, marking a significant advancement in the field of construction equipment monitoring. [ABSTRACT FROM AUTHOR]

The decipherment of ancient Chinese scripts, such as oracle bone and bronze inscriptions, holds immense significance for understanding ancient Chinese history, culture, and civilization. Despite substantial progress in recognizing oracle bone script, research on the overall recognition of ancient Chinese characters remains somewhat lacking. To tackle this issue, we pioneered the construction of a large-scale image dataset comprising 9233 distinct ancient Chinese characters sourced from images obtained through archaeological excavations. We propose the first model for recognizing the common ancient Chinese characters. This model consists of four stages with Linear Embedding and Swin-Transformer blocks, each supplemented by a CoT Block to enhance local feature extraction. We also advocate for an enhancement strategy, which involves two steps: firstly, conducting adaptive data enhancement on the original data, and secondly, randomly resampling the data. The experimental results, with a top-one accuracy of 87.25% and a top-five accuracy of 95.81%, demonstrate that our proposed method achieves remarkable performance. Furthermore, through the visualizing of model attention, it can be observed that the proposed model, trained on a large number of images, is able to capture the morphological characteristics of ancient Chinese characters to a certain extent. [ABSTRACT FROM AUTHOR]

In real‐world scenarios, pedestrian images often suffer from occlusion, where certain body features become invisible, making it challenging for existing methods to accurately identify pedestrians with the same ID. Traditional approaches typically focus on matching only the visible body parts, which can lead to misalignment when the occlusion patterns vary. To address this issue and alleviate misalignment in occluded pedestrian images, the authors propose a novel framework called body topology information generation and matching. The framework consists of two main modules: the body topology information generation module and the body topology information matching module. The body topology information generation module employs an adaptive detection mechanism and capsule generative adversarial network to restore a holistic pedestrian image while preserving the body topology information. The body topology information matching module leverages the restored holistic image from body topology information generation to overcome spatial misalignment and utilises cosine distance as the similarity measure for matching. By combining the body topology information generation and body topology information matching modules, the authors achieve consistency in the body topology information features of pedestrian images, ranging from restoration to retrieval. Extensive experiments are conducted on both holistic person re‐identification datasets (Market‐1501, DukeMTMC‐ReID) and occluded person re‐identification datasets (Occluded‐DukeMTMC, Occluded‐ReID). The results demonstrate the superior performance of the authors proposed model, and visualisations of the generation and matching modules are provided to illustrate their effectiveness. Furthermore, an ablation study is conducted to validate the contributions of the proposed framework. [ABSTRACT FROM AUTHOR]

Semantic segmentation from a three‐dimensional point cloud is vital in autonomous driving, computer vision, and augmented reality. However, current semantic segmentation does not effectively use the point cloud's local geometric features and contextual information, essential for improving segmentation accuracy. A semantic segmentation network that uses local feature fusion and a multilayer attention mechanism is proposed to address these challenges. Specifically, the authors designed a local feature fusion module to encode the geometric and feature information separately, which fully leverages the point cloud's feature perception and geometric structure representation. Furthermore, the authors designed a multilayer attention pooling module consisting of local attention pooling and cascade attention pooling to extract contextual information. Local attention pooling is used to learn local neighbourhood information, and cascade attention pooling captures contextual information from deeper local neighbourhoods. Finally, an enhanced feature representation of important information is obtained by aggregating the features from the two deep attention pooling methods. Extensive experiments on large‐scale point‐cloud datasets Stanford 3D large‐scale indoor spaces and SemanticKITTI indicate that authors network shows excellent advantages over existing representative methods regarding local geometric feature description and global contextual relationships. [ABSTRACT FROM AUTHOR]

Aircraft coating process has been an important part in manufacturing process of modern aviation products. For coating defect detection, the manual observation with naked eyes is usually utilized, which leads to low production efficiency. In this paper, the authors propose the improved YOLOv5‐OBB with the channel‐spatial attention block (CSAB), feature pyramid non‐local module (FPNM) and structured sparsity slimming criterion (SSSC). The CSAB can pay more attention to effective channel information features from the channel dimension and the target information area from the spatial dimension. The effective non‐local module called FPNM is proposed to further improve the detection accuracy. The authors utilize the oriented bounding boxes (OBB) to reduce redundant background information for coating defect detection. In addition, the SSSC is proposed to achieve network slimming and trade‐off between the efficiency and accuracy. The experimental results on several datasets demonstrate the effectiveness of the authors' scheme, which achieves superior performance. [ABSTRACT FROM AUTHOR]

In recent years, various studies have been conducted on SVMs and their applications in different area. They have been developed significantly in many areas. SVM is one of the most robust classification and regression algorithms that plays a significant role in pattern recognition. However, SVM has not been developed significantly in some areas like large-scale datasets, unbalanced datasets, and multiclass classification. Efficient SVM training in large-scale datasets is of great importance in the big data era. However, as the number of samples increases, the time and memory required to train SVM increase, making SVM impractical even for a medium-sized problem. With the emergence of big data, this problem becomes more significant. This paper presents a novel distributed method for SVM training in which a very small subset of training samples is used for classification, which reduces the problem size and thus the required memory and computational resources. The solution of this problem almost converges to standard SVM. This method includes three steps: first, detecting a subset of distributed training samples, second, creating local models of SVM and obtaining partial vectors, and finally combining the partial vectors and obtaining the global vector and the final model. In addition, the datasets which suffer from unbalanced number of samples and tend to the majority class, the proposed method balances the samples of the two classes and it can be used in unbalanced datasets. The empirical results show that using this method is efficient for large-scale problems. [ABSTRACT FROM AUTHOR]

Movie scene segmentation aims to automatically segment a movie into multiple story units, i.e., scenes, each of which is a series of semantically coherent and time-continual shots. Previous methods have continued efforts on shot semantic association, but few take into account the impact of different semantics on foreground characters and background scenes in movie shots. In particular, the background scene in the shot can adversely affect scene boundary classification. Motivated by the fact that it is the characters who drive the plot development of a movie scene, we build a Character Attention Network (CANet) to detect movie scene boundaries in a character-centric fashion. To eliminate the background clutter, we extract multi-view character semantics for each shot in terms of human bodies and faces. Furthermore, we equip our CANet with two stages of character attention. The first is Masked Shot Attention (MSA) through selective self-attention over similar temporal contexts from multi-view character semantics to yield an enhanced omni-view shot representation, by which the CANet can better handle the variations of characters in pose and appearance. The second is Key Character Attention (KCA) through temporal-aware attention on character reappearances for Bidirectional Long Short-Term Memory (Bi-LSTM) feature association so that linking shots can be focused on those with recurring key characters. We encourage the proposed CANet in learning boundary-discriminative shot features. Specifically, we formulate a Boundary-Aware circle Loss (BAL) to push far apart CANet-features between adjacent scenes, which is also coupled with the cross-entropy loss to drive CANet-features sensitive to scene boundaries. Experimental results on the MovieNet-SSeg and OVSD datasets show that our method achieves superior performance in temporal scene segmentation compared with state-of-the-art methods. [ABSTRACT FROM AUTHOR]