Search

Handwriting is a biometric behavioral characteristic with evident individual distinctiveness. With the rise of the deep learning trend and demands for forensic identification, handwriting identification has become one of the focal points of research in the field of pattern recognition. Research in handwriting identification for major global languages has matured. However, in China, there is limited attention in the field of writer identification for minority languages such as Mongolian, making it challenging to resolve criminal cases involving handwriting issues. This paper initiates an initial exploration of Mongolian handwriting identification by constructing a structurally simple convolutional neural network. This convolutional neural network, consisting of 12 convolution operations and designed for Mongolian handwriting identification, is referred to as MWInet-12. In this paper, the model evaluation experiments were conducted using a dataset comprising 156,372 samples contributed by 125 writers from the MOLHW dataset. The dataset was divided into training, validation, and test sets in an 8:1:1 ratio. The final results of the experiments reveal impressive accuracy on the test set, achieving a top-1 accuracy of 89.60% and a top-5 accuracy of 97.53%. Furthermore, through comparative experiments involving Resnet50, Fragnet, GRRNN, VGG16, and VGG19 models, this paper establishes that the proposed model yields the most favorable results for Mongolian handwriting identification. The exploratory research on Mongolian handwriting identification in this paper contributes to increasing awareness of information processing for minority languages. It aids in advancing research on classifying writers of Mongolian historical texts and provides technical support for judicial authentication involving handwriting issues. [ABSTRACT FROM AUTHOR]

Defect pattern analysis of wafer bin maps (WBMs) is an important means of identifying process problems. Recently, automated analysis methods using machine learning or deep learning have been studied as alternatives to manual classification by engineers. In this paper, we propose a method to improve the feature extraction performance of defect patterns by transforming the polar coordinate system instead of the existing WBM image input. To reduce the variability of the location representation, defect patterns in the Cartesian coordinate system, where the location of the distributed defect die is not constant, were converted to a polar coordinate system. The CNN classifier, which uses polar coordinate transformed input, achieved a classification accuracy of 91.3%, which is 4.8% better than the existing WBM image-based CNN classifier. Additionally, a tree-structured classifier model that sequentially connects binary classifiers achieved a classification accuracy of 94%. The method proposed in this paper is also applicable to the defect pattern classification of WBMs consisting of different die sizes than the training data. Finally, the paper proposes an automated pattern classification method that uses individual classifiers to learn defect types and then applies ensemble techniques for multiple defect pattern classification. This method is expected to reduce labor, time, and cost and enable objective labeling instead of relying on subjective judgments of engineers. [ABSTRACT FROM AUTHOR]

Mental workload is an important predisposing factor for mental illnesses such as depression and is closely related to individual mental health. However, the suboptimal accuracy of utilizing photoplethysmography (PPG) exclusively for mental workload classification has constrained its application within pertinent professional domains. To this end, this paper proposes a signal processing method that combines continuous wavelet transform (CWT) and cardiopulmonary coupling mapping (CPC) to classify mental load via a convolutional neural network (ResAttNet). The method reflects changes in mental workload, as assessed by changes in the association between heart rate variability and respiration. In this paper, the strengths and weaknesses of this method are compared with other traditional psychological workload monitoring methods, such as heart rate variability (HRV), and its validation is performed on the publicly available dataset MAUS. The experiments show that the method is significantly better than previous machine learning methods based on heart rate variability correlation. Meanwhile, the accuracy of the method proposed in this paper reaches 80.5%, which is 6.2% higher than in previous studies. It is comparable to the result of 82.4% for the ECG-based mental workload monitoring system. Therefore, the method of combining CWT and CPC has considerable potential and provides new ideas for mental workload classification. [ABSTRACT FROM AUTHOR]

Convenient and accurate blood pressure (BP) measurement is of great importance in both clinical and daily life. Although deep learning (DL) can achieve cuff-less BP measurement based on Photoplethysmography (PPG), the performance of DL is affected by few-shot data. Data augmentation becomes an effective way to enhance the size of the training data. In this paper, we use cropping, flipping, DTW barycentric averaging (DBA), generative adversarial network (GAN) and variational auto-encoder (VAE) for the data augmentation of PPG. Furthermore, a PE–CNN–GRU model is designed for cuff-less BP measurement applying position encoding (PE), convolutional neural networks (CNNs) and gated recurrent unit (GRU) networks. Experiment results based on real-life datasets show that VAE is the most suitable method for PPG data augmentation, which can reduce the mean absolute error (MAE) of PE–CNN–GRU when measuring systolic blood pressure (SBP) and diastolic blood pressure (DBP) by 18.80% and 19.84%. After the data augmentation of PPG, PE–CNN–GRU achieves accurate and cuff-less BP measurement, thus providing convenient support for preventing cardiovascular diseases. [ABSTRACT FROM AUTHOR]

In recent decades, with the ever-growing scale of video data, near-duplicate videos continue to emerge. Data quality issues caused by near-duplicate videos are becoming more and more prominent, which has affected the application of normal videos. Although current studies on near-duplicate video detection can help uncover data quality issues for videos, they still lack a process of automatic merging for the video data represented by high-dimensional features, which makes it difficult to automatically clean the near-duplicate videos to improve data quality for video datasets. At present, there are few studies on near-duplicate video data cleaning. The existing studies have the sensitive problems of video data orderliness and initial clustering centers under a condition that prior distribution is unknown, which seriously affects the accuracy of near-duplicate video data cleaning. To address the above issues, an automatic near-duplicate video data cleaning method based on a consistent feature hash ring is proposed in this paper. First, a residual network with convolutional block attention modules, a long short-term memory deep network, and an attention model are integrated to construct an RCLA deep network with the multi-head attention mechanism to extract spatiotemporal features of video data. Then, a consistent feature hash ring is constructed, which can effectively alleviate the sensitivity of video data orderliness while providing a condition of near-duplicate video merging. To reduce the sensitivity of the initial cluster centers to the results of near-duplicate video cleansing, an optimized feature distance-means clustering algorithm is constructed by utilizing a mountain peak function on a consistent feature hash ring, which can implement automatic cleaning of near-duplicate video data. Finally, experiments are conducted based on a commonly used dataset named CC_WEB_VIDEO and a coal mining video dataset. Compared with some existing studies, simulation results demonstrate the performance of the proposed method. [ABSTRACT FROM AUTHOR]

Reinforcement learning (RL) techniques have emerged as powerful tools for optimizing energy systems, offering the potential to enhance efficiency, reliability, and sustainability. This review paper provides a comprehensive examination of the applications of RL in the field of energy system optimization, spanning various domains such as energy management, grid control, and renewable energy integration. Beginning with an overview of RL fundamentals, the paper explores recent advancements in RL algorithms and their adaptation to address the unique challenges of energy system optimization. Case studies and real-world applications demonstrate the efficacy of RL-based approaches in improving energy efficiency, reducing costs, and mitigating environmental impacts. Furthermore, the paper discusses future directions and challenges, including scalability, interpretability, and integration with domain knowledge. By synthesizing the latest research findings and identifying key areas for further investigation, this paper aims to inform and inspire future research endeavors in the intersection of reinforcement learning and energy system optimization. [ABSTRACT FROM AUTHOR]

Detecting and identifying small infrared targets has always been a crucial technology for many applications. To address the low accuracy, high false-alarm rate, and poor environmental adaptability that commonly exist in infrared target detection methods, this paper proposes a composite infrared dim and small target detection model called USES-Net, which combines the target prior knowledge and conventional data-driven deep learning networks to make use of both labeled data and the domain knowledge. Based on the typical encoder–decoder structure, USES-Net firstly introduces the self-attention mechanism of Swin Transformer to replace the universal convolution kernel at the encoder end. This helps to extract potential features related to dim, small targets in a larger receptive field. In addition, USES-Net includes an embedded patch-based contrast learning module (EPCLM) to integrate the spatial distribution of the target as a knowledge prior in the training network model. This guides the training process of the constrained network model with clear physical interpretability. Finally, USES-Net also designs a bottom-up cross-layer feature fusion module (AFM) as the decoder of the network, and a data-slicing-aided enhancement and inference method based on Slicing Aided Hyper Inference (SAHI) is utilized to further improve the model's detection accuracy. An experimental comparative analysis shows that USES-Net achieves the best results on three typical infrared weak-target datasets: NUAA-SIRST, NUDT-SIRST, and IRSTD-1K. The results of the target segmentation are complete and sufficient, which demonstrates the validity and practicality of the proposed method in comparison to others. [ABSTRACT FROM AUTHOR]

The swift embrace of Industry 4.0 paradigms has led to the growing convergence of Information Technology (IT) networks and Operational Technology (OT) networks. Traditionally isolated on air-gapped and fully trusted networks, OT networks are now becoming more interconnected with IT networks due to the advancement and applications of IoT. This expanded attack surface has led to vulnerabilities in Cyber–Physical Systems (CPSs), resulting in increasingly frequent compromises with substantial economic and life safety repercussions. The existing methods for the anomaly detection of security threats typically use simple threshold-based strategies or apply Machine Learning (ML) algorithms to historical data for the prediction of future anomalies. However, due to the high levels of heterogeneity across different CPS environments, minimizing the opportunities for transfer learning, and the scarcity of real-world data for training, the existing ML-based anomaly detection techniques suffer from a poor predictive performance. This paper introduces a hybrid anomaly detection approach designed to identify threats to CPSs by combining the signature-based anomaly detection typically utilized in IT networks, the threshold-based anomaly detection typically utilized in OT networks, and behavioural-based anomaly detection using Ensemble Learning (EL), which leverages the strengths of multiple ML algorithms against the same dataset to increase the accuracy. Multiple public research datasets were used to validate the proposed approach, with the hybrid methodology employing a divide-and-conquer strategy to offload the detection of certain cyber threats to computationally inexpensive signature-based and threshold-based methods using domain knowledge to minimize the size of the behavioural-based data needed for ML model training, thus achieving a higher accuracy over a reduced timeframe. The experimental results showed accuracy improvements of 4–7% over those of the conventional ML classifiers in performing anomaly detection across multiple datasets, which is particularly important to the operators of CPS environments due to the high financial and life safety costs associated with interruptions to system availability. [ABSTRACT FROM AUTHOR]

Bit-serial neural network accelerators address the growing need for compact and energy-efficient deep learning tools. Traditional neural network accelerators, while effective, often grapple with issues of size, power consumption, and versatility in handling a variety of computational tasks. To counter these challenges, this paper introduces an approach that hinges on the integration of bit-serial processing with advanced dataflow techniques and architectural optimizations. Central to this approach is a column-buffering (CB) dataflow, which significantly reduces access and movement requirements for the input feature map (IFM), thereby enhancing efficiency. Moreover, a simplified quantization process effectively eliminates biases, streamlining the overall computation process. Furthermore, this paper presents a meticulously designed LeNet-5 accelerator leveraging a convolutional layer processing element array (CL PEA) architecture incorporating an improved bit-serial multiply–accumulate unit (MAC). Empirically, our work demonstrates superior performance in terms of frequency, chip area, and power consumption compared to current state-of-the-art ASIC designs. Specifically, our design utilizes fewer hardware resources to implement a complete accelerator, achieving a high performance of 7.87 GOPS on a Xilinx Kintex-7 FPGA with a brief processing time of 284.13 μs. The results affirm that our design is exceptionally suited for applications requiring compact, low-power, and real-time solutions. [ABSTRACT FROM AUTHOR]

To address the issue of low accuracy in detecting malicious program behaviors in new power system edge-side applications, we present a detection model based on API call sequences that combines rule matching and deep learning techniques in this paper. We first use the PrefixSpan algorithm to mine frequent API call sequences in different threads of the same program within a malicious program dataset to create a rule base for malicious behavior sequences. The API call sequences to be examined are then matched using the malicious behavior sequence matching model, and those that do not match are fed into the TextCNN deep learning detection model for additional detection. The two models collaborate to accomplish program behavior detection. Experimental results demonstrate that the proposed detection model can effectively identify malicious samples and discern malicious program behaviors. [ABSTRACT FROM AUTHOR]