检索结果-内蒙古大学图书馆

Multi-Scale Time Series Segmentation Network Based on Eddy Current Testing for Detecting Surface Metal Defects

IEEE/CAA Journal of Automatica Sinica 2025年第3期12卷 528-538页

作者： Xiaorui Li Xiaojuan Ban Haoran Qiao Zhaolin Yuan Hong-Ning Dai Chao Yao Yu Guo Mohammad S.Obaidat George Q.Huang the School of Intelligence Science and Technology University of Science and Technology Beijing the Beijing Advanced Innovation Center for Materials Genome Engineering the Key Laboratory of Intelligent Bionic Unmanned Systems and the Institute of Materials Intelligent Technology Liaoning Academy of Materials IEEE the Department of Computer Science Hong Kong Baptist University the School of Computer and Communication Engineering Key Laboratory of Advanced Materials and Devices for Post-Moore Chips Ministry of Education University of Science and Technology Beijing the Beijing Advanced Innovation Center for Materials Genome Engineering University of Science and Technology Beijing the School of Computer and Communication Engineering University of Science and Technology Beijing the King Abdullah Ⅱ School of Information Technology The University of Jordan the Department of Computational Intelligence the School of Computing SRM University the School of Engineering The Amity University The Hong Kong Polytechnic University

In high-risk industrial environments like nuclear power plants, precise defect identification and localization are essential for maintaining production stability and safety. However, the complexity of such a harsh environment leads to significant variations in the shape and size of the defects. To address this challenge, we propose the multivariate time series segmentation network(MSSN), which adopts a multiscale convolutional network with multi-stage and depth-separable convolutions for efficient feature extraction through variable-length templates. To tackle the classification difficulty caused by structural signal variance, MSSN employs logarithmic normalization to adjust instance distributions. Furthermore, it integrates classification with smoothing loss functions to accurately identify defect segments amid similar structural and defect signal subsequences. Our algorithm evaluated on both the Mackey-Glass dataset and industrial dataset achieves over 95% localization and demonstrates the capture capability on the synthetic dataset. In a nuclear plant's heat transfer tube dataset, it captures 90% of defect instances with75% middle localization F1 score.

关键词： Eddy current testing nondestructive testing semantic segmentation time series analysis

来源：评论

学校读者我要写书评

暂无评论

An effective object detection via BS2ResNet and LTK-Bi-LSTM

引用

Multimedia Tools and Applications 2025年 1-20页

作者： V, Arulalan V, Premanand Kumar, Dhananjay Department of Computing Technologies SRM Institute of Science and Technology Chengalpattu District Tamil Nadu Kattankulathur603203 India Department of Computer Science and Engineering Vellore Institute Technology Chennai Campus Tamil Nadu Chennai India Department of Information Technology Anna University MIT Campus Tamil Nadu Chennai India

Video analytics faces complex challenges in object detection and classification. Deep learning based approaches have achieved remarkable success in past decade. However, existing object identification models that utilize backbone’s core features still present challenges due to their lack of semantic information. To address these issues, a novel object detection and classification framework utilizing Batch normalization and Softswish activation adapted ResNet (BS2ResNet) and Logistic Tanh Kaiming Bi-directional Long Short Term Memory (LTK-Bi-LSTM) techniques was proposed. The framework employs frame conversion, noise removal, and contrast elevation during frame pre-processing, followed by background subtraction using the Supreme Distance-centered Fuzzy C-Means (SD-FCM) clustering algorithm, and edge detection using the Hyperbolic Tangent Kernel Canny Edge Detector (HTKCED). BS2ResNet is then used for object detection, and features are extracted and passed to the LTK-Bi-LSTM neural network for object classification. The proposed system was found to improve object detection and classification accuracy, outperforming existing techniques. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

LoRa WAN Architecture with Collusion Management and Collusion Avoidances Schemes in MAC Layer

LoRa WAN Architecture with Collusion Management and Collusio...

引用

International Conference on computing and Machine Learning, CML 2024

作者： Saravanan, N. Kayalvizhi, R. Department of Networking and Communications School of Computing College of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Chengalpattu Tamil Nadu India

ISBN: (纸本)9789819778386

Long Range (LoRa) is the most widely used technology for significant attention in recent years due to their ability to enable long-range, low-power connectivity, making it an ideal communication system for Internet of Things applications. LoRa radio is considered as the physical layer, MAC layer which was specifically developed for low power IoT devices. LoRa is one of the technologies in LPWAN, which demonstrated remarkable potential in various domains used for IoT applications. This research paper presents a comprehensive study of LoRa technology, covering its LoRa principles (Physical Layer, MAC Layer, Network layer) Communication Mechanism, Application usage. Through an in-depth analysis of the key features and advantages of LoRa, this paper aims to provide a holistic understanding of the technology and its potential implications for the future of IoT in Collusion management and collision avoidance. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Information management

来源：评论

学校读者我要写书评

暂无评论

Recognition of American sign language using modified deep residual CNN with modified canny edge segmentation

引用

Multimedia Tools and Applications 2025年 1-28页

作者： Hariharan, U. Devarajan, Devikanniga Kumar, P. Santhosh Rajkumar, K. Meena, M. Akilan, T. Department of Computer Science and Engineering Apex Institute of Technology Chandigarh University Punjab Mohali140413 India Department of Computer Science and Engineering GITAM University Karnataka Bengaluru561203 India Department of Information Technology Faculty of Engineering and technology SRM Institute of Science and Technology Ramapuram Tamil Nadu Chennai600089 India Department of Data Science and Business Systems School of Computing Faculty of Engineering and Technology SRM Institute of Science and Technology Kattankulathur Tamil Nadu Chennai603203 India Department of Computer Science and Engineering Sri Sairam Engineering College Tamil Nadu Tambaram602109 India Research Coordinator Dept. of IQAC Amity University Kolkata West Bengal 700135 India

American Sign Language (ASL) recognition aims to recognize hand gestures, and it is a crucial solution to communicating between the deaf community and hearing people. However, existing sign language recognition algorithms still have some drawbacks, such as difficulty recognizing hand movements and low recognition accuracy for most sign language recognition. A Modified Convolutional Neural Network (MCNN) deep residual 101 classifier-based American Sign Language identification system has been developed to address this problem. There are three main parts presents in the method. The first part is pre-processing the images to remove the noise, enhance the contrast of the picture, adjust the contrast level and smoothen the picture using various filters. The second part is the segmentation, and it’s used to partition the image using a modified canny edge detection method by removing all weak edges present in the image. Finally, classification will be done using the Modified CNN deep residual 101 classifiers. By classifying the image, the American Sign Language is accurately identified. This process is conducted through images. The outcome shows that the suggested approach has a 0.97% accuracy and a 0.05% False Positive Rate. Other CNN architectures such as resNet 50 and resNet 18 reached 0.90% and 0.80% accuracy, respectively, which is lower than our proposed method. In addition, the resNet 101 classifier effectively recognizes the difficult hand gestures through the image data. It obtains high recognition accuracy for 36 signs is 0 to 9 numbers and a to z alphabets from American Sign Language. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Real-Time On-Site Ground-Motion Prediction Using ResNet

引用

Seismological Research Letters 2025年第3期96卷 1834-1845页

作者： Yang, Tzu-Yi Chen, Da-Yi Chin, Tai-Lin Mobile Computing and Networking Lab Department of Computer Science and Information Engineering National Taiwan University of Science and Technology Taipei Taiwan Seismological Center of Central Weather Administration Taipei Taiwan

On-site warnings can decrease the range of late alert zone during earthquakes. This study develops a deep learning model to predict whether the maximum peak ground acceleration at a station exceeds 25 Gal based on the waveform of the initial P-wave. A ResNet architecture model is developed to address the degradation problem in multilayered models, thereby enhancing performance. The model exhibits high performance for a test dataset, achieving F1-scores higher than 90% over various time windows, and outperforms the traditional progressive displacement value (Pdv) method and newer convolutional neural network (CNN) methods. Ground motion of two actual earthquakes are applied to test the model and validate its prediction capabilities, and the results indicate that the proposed model has higher accuracy and speed than Pdv and CNN methods. In addition, the model is integrated with the Earthworm earthquake early warning system for real-time waveform analysis and prediction, extending the model’s applicability beyond experimental stages to online systems. The proposed method is also tested using data from earthquakes that occurred in March 2023. © Seismological Society of America.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A deep learning approach for recognizing and solving handwritten mathematical equations

引用

Neural computing and Applications 2025年第14期37卷 8759-8772页

作者： Yadav, Pavinder Shantilal, Sankhala Bhavik Kumar, Vipin Sihag, Parveen Sharma, Pawan Kumar Rana, Pankaj Department of Mathematics and Scientific Computing National Institute of Technology HP Hamirpur India Department of Civil Engineering Chandigarh University Punjab Mohali India Department of Physical Sciences Baba Mastnath University Asthal Bohar Haryana Rohtak India

Recognizing handwritten equations is considerably more challenging than identifying printed ones, primarily due to variations in writing styles. Even though there have been recent improvements in handwriting recognition for things like equation solvers and automation, it is still hard for machines to understand a handwritten mathematical expression (HME). This paper presents a deep learning model for recognizing and solving handwritten equations, as well as its graphical user interface (GUI). A manually constructed large dataset comprising approximately 66k images of digits, mathematical symbols, and characters is utilized, with a fine-tuned InceptionV3 model employed for image classification and segmentation. The system is designed to extract handwritten equations from an image, perform character recognition, and generate the roots of the polynomial equation up to degree four. According to the findings, the trained model achieves remarkable accuracy in recognizing and solving handwritten equations, with over 94% accuracy on the training set. By eliminating the possibility of human error, automatic handwritten equation solvers provide more accurate results than manual methods. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

An Iterative PRISMA Review of GAN Models for Image Processing, Medical Diagnosis, and Network Security

引用

Computers, Materials & Continua 2025年第2期82卷 1757-1810页

作者： Uddagiri Sirisha Chanumolu Kiran Kumar Sujatha Canavoy Narahari Parvathaneni Naga Srinivasu Department of Computer Science and Engineering Prasad V Potluri Siddhartha Institute of TechnologyVijayawada520007India Department of Computer Science and Engineering Koneru Lakshmaiah Education FoundationVaddeswaram522302India Department of Electronics and Communication Engineering Sreenidhi Institute of Science and TechnologyHyderabad501301India Amrita School of Computing Amrita Vishwa VidyapeethamAmaravati522503India Department of Teleinformatics Engineering Federal University of CearáFortaleza60455-970Brazil Inti International University Nilai71800Malaysia

The growing spectrum of Generative Adversarial Network (GAN) applications in medical imaging, cyber security, data augmentation, and the field of remote sensing tasks necessitate a sharp spike in the criticality of review of Generative Adversarial Networks. Earlier reviews that targeted reviewing certain architecture of the GAN or emphasizing a specific application-oriented area have done so in a narrow spirit and lacked the systematic comparative analysis of the models’ performance metrics. Numerous reviews do not apply standardized frameworks, showing gaps in the efficiency evaluation of GANs, training stability, and suitability for specific tasks. In this work, a systemic review of GAN models using the PRISMA framework is developed in detail to fill the gap by structurally evaluating GAN architectures. A wide variety of GAN models have been discussed in this review, starting from the basic Conditional GAN, Wasserstein GAN, and Deep Convolutional GAN, and have gone down to many specialized models, such as EVAGAN, FCGAN, and SIF-GAN, for different applications across various domains like fault diagnosis, network security, medical imaging, and image segmentation. The PRISMA methodology systematically filters relevant studies by inclusion and exclusion criteria to ensure transparency and replicability in the review process. Hence, all models are assessed relative to specific performance metrics such as accuracy, stability, and computational efficiency. There are multiple benefits to using the PRISMA approach in this setup. Not only does this help in finding optimal models suitable for various applications, but it also provides an explicit framework for comparing GAN performance. In addition to this, diverse types of GAN are included to ensure a comprehensive view of the state-of-the-art techniques. This work is essential not only in terms of its result but also because it guides the direction of future research by pinpointing which types of applications require some

关键词： GAN CGAN WGAN DCGAN image analysis

来源：评论

学校读者我要写书评

暂无评论

PneuX-Net: An Enhanced Feature Extraction and Transformation Approach for Pneumonia Detection in X-Ray Images

引用

IEEE Access 2025年 13卷 84024-84037页

作者： Munir, Kashif Usama Tanveer, Muhammad Alyamani, Hasan J. Bermak, Amine Ur Rehman, Atiq Khwaja Fareed University of Engineering and Information Technology Institute of Information Technology Rahim Yar Khan 64200 Pakistan King Abdulaziz University Faculty of Computing and Information Technology in Rabigh Department of Information Systems Jeddah 21589 Saudi Arabia Hamad Bin Khalifa University Division of Information and Computing Technology College of Science and Engineering Doha Qatar

Pneumonia is a critical respiratory condition characterized by inflammation in the alveoli, leading to impaired gas exchange and severe respiratory distress. It poses a significant threat to high-risk populations, including neonates, geriatric patients, and immunocompromised individuals. Early and precise detection is crucial to optimizing treatment strategies and reducing morbidity and mortality rates. To address this problem, we propose PneuX-Net, an ensemble-based feature extraction framework that integrates multiple machine learning (ML) models Random Forest (RF), Gaussian Naïve Bayes (GNB) and K-Nearest Centroid (KNC). Random Forest (RF) builds multiple decision trees to identify complex, non-linear patterns within the feature space. Gaussian Naïve Bayes (GNB) utilizes a probabilistic framework based on Bayes’ theorem to manage uncertainty in feature distributions. K-Nearest Centroid (KNC) improves class distinction by clustering feature vectors according to their proximity to centroids. The ensemble methodology harnesses the complementary strengths of these models, improving feature representation and mitigating overfitting, a prevalent issue in white-box models. To validate the effectiveness of PneuX-Net, we conduct extensive experimentation using a 10-fold cross-validation approach. Our results demonstrate that the PneuX-Net+KNC model achieves a remarkable 99.91% accuracy, highlighting its robustness and reliability in pneumonia classification tasks. This ensemble-driven methodology underscores the potential of machine learning in augmenting clinical decision-making by providing an accurate and automated pneumonia detection framework. © 2013 IEEE.

关键词： healthcare AI machine learning medical diagnosis pneumonia detection PneuX-Net

来源：评论

学校读者我要写书评

暂无评论

A novel context-sensitive attitude entropy-based multiclass segmentation method for brain MR images using enhanced flow directional algorithm

引用

Multimedia Tools and Applications 2025年第15期84卷 14759-14803页

作者： Naik, Manoj Kumar Jena, Bibekananda Panda, Rutuparna Wunnava, Aneesh Abraham, Ajith Faculty of Engineering and Technology Siksha O Anusandhan Odisha Bhubaneswar751030 India Department of Electronics and Communication Engineering Anil Neerukonda Institute of Technology & Sciences Andhra Pradesh Sangivalasa Visakhapatnam531162 India Department of Electronics and Telecommunication Engineering Veer Surendra Sai University of Technology Odisha Burla768018 India Faculty of Computing and Mathematical Sciences Bennett University Greater Noida201310 India

Computer vision techniques, aided by artificial intelligence, in medical image analysis, are currently encouraged for precise and speedy diagnosis of brain disorders. The brain MR image shows an intensity variation, which attracts multiclass segmentation for a meaningful representation. The existing entropy-based methods use the image histogram. Nevertheless, the 1D histogram-based technique has a low complexity. However, it lacks the contextual information. Further, these entropy-based methods suffer low accuracy when the gray value distribution is non-uniform. To facilitate a reliable and efficient thresholding process, a novel multiclass segmentation objective function is investigated along with an efficient optimizer to obtain the optimal threshold values. To bridge this research gap, we explore a context-sensitive attitude entropy (CSAE) based multiclass segmentation objective function, an energy curve is used together with the human decision-making ability. Context-sensitive attitude entropy (CSAE), in contrast to classical entropy, which considers each pixel separately, is a measure of uncertainty or randomness in an image considering the spatial relationships and context between adjacent pixels. This can be achieved by calculating the attitude entropy from the energy curve instead of the image’s histogram. To maximize the CSAE, we suggest a new enhanced flow directional algorithm (EFDA), which improves the flow directional algorithm (FDA) by adding randomization to both the velocity and the non-performer agents. The experimental effort begins with the EFDA performance validation utilising 23 well-known classical and 10 CEC-C06 2019 benchmark functions, which produce better results than cutting-edge optimizers. The EFDA’s supremacy over other optimizers is ensured by Friedman’s mean rank and Wilcoxon’s mean rank test. The suggested CSAE-EFDA method is tested using T2-weighted brain images from Harvard Medical School’s AANLIB database and Brain MR images for t

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Enhancing Gallium Arsenide Semiconductor Chip Quality Classification through AI Algorithms

引用

Sensors and Materials 2025年第5期37卷 1785-1807页

作者： Luo, Guofeng Lee, Hsiao-Yi Wang, Kenvi Department of Electrical Engineering National Kaohsiung University of Science and Technology Kaohsiung 80778 Taiwan Department of Computing and Mathematics Liverpool John Moores University United Kingdom

Deep learning algorithms, such as those employed by CHATGPT/artificial neural networks, have served as transformative catalysts across various sectors, significantly impacting our daily lives. Within the realm of semiconductor processes, the metric of yield stands as a crucial performance benchmark directly influencing the sustained success of enterprises. Among batches of chips, inherent disparities in quality persist, with each chip grade necessitating an appropriate operational environment. To preempt the pitfalls of customer returns or factory repairs, thereby mitigating manufacturing costs, in this study, we embark on a comprehensive exploration utilizing AI algorithms for quality classification. In pursuit of bolstering the competitive edge of semiconductor firms, the refinement of gallium arsenide chip quality classification to augment yield emerges as a paramount concern. Moreover, in this research endeavor, we seek to discern the physical parameters underlying GaAs defect products, offering pivotal insights to enhance the manufacturing process—an aspect poised to be a pivotal focal point. © MYU K.K.

关键词： accuracy AI artificial neural network gallium arsenide machine learning

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：