检索结果-内蒙古大学图书馆

Optimized Convolutional Neural Network at the IoT edge for image detection using pruning and quantization

Multimedia Tools and Applications 2025年第9期84卷 5435-5455页

作者： Naveen, Soumyalatha Kounte, Manjunath R Department of Computer Science and Engineering Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal India School of Electronics and Communication Engineering REVA University Bangalore India

Most real-time computer vision applications heavily rely on Convolutional Neural Network (CNN) based models, for image classification and recognition. Due to the computationally and memory-intensive nature of the CNN model, it’s challenging to deploy on resource-constrained Internet of Things (IoT) devices to enable Edge intelligence for real-time decision-making. Edge intelligence requires minimum inference latency, memory footprint, and energy-efficient model. This work aims to develop an energy-efficient deep learning accelerator using a 3-stage pipeline: Training, Weight-pruning, and Quantization to reduce the model size and optimize the resources. First, we employ YOLOv3, a CNN architecture to detect objects in an image on the trained data. In addition, a sparse network of YOLO has been created by using pruning, which helps to improve the network’s performance and efficiency by reducing the computational requirements. Finally, we utilize 8-bit quantization to reduce the precision of the weights and activations, in a neural network. The evaluation of our proposed model shows that combining pruning and 8-bit quantization improves the efficiency and performance of the model. While pruning shows a decline of 80.39% in model parameters. The combination of 8-bit quantization results in an improvement in inference latency by 22.72% compared to existing SQuantization approach and a reduction of energy consumption by 29.41%. © The Author(s) 2024.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

"Survey and analysis of blockchain technologies with respect to: properties, algorithms, architecture, models, evolution and framework"

引用

Multimedia Tools and Applications 2025年第14期84卷 13571-13615页

作者： Tiwari, Nikita Ranjan, Prakash Biswal, Pradeep Kumar Barde, Chetan Sinha, Neha Department of Computer Science and Engineering Indian Institute of Information Technology Bhagalpur Bhagalpur India Department of Electronics and Communication Engineering Indian Institute of Information Technology Bhagalpur Bhagalpur India

The growing realm of blockchain technology has captivated researchers and practitioners alike with its promise of decentralized, secure, and transparent transactions. This paper presents a comprehensive survey and analysis of various blockchain technologies, focusing on their fundamental properties, architectural designs, operational models, evolutionary trends, and overarching frameworks. Through an extensive review of existing literature and empirical studies, this research synthesizes the multifaceted landscape of blockchain, elucidating key concepts such as consensus mechanisms, smart contracts, scalability, interoperability, and privacy protocols. By examining the evolution of blockchain from its inception to its current state and beyond, this paper delineates the trajectory of technological advancements and identifies emerging paradigms. Furthermore, it critically evaluates the strengths and limitations of diverse blockchain architectures and models, offering insights into their applicability across different domains and use cases. Finally, this study culminates in the proposal of a comprehensive framework that encapsulates the essential elements necessary for understanding, implementing, and harnessing the transformative potential of blockchain technologies. Through this holistic analysis, this research contributes to a deeper understanding of blockchain ecosystems and provides valuable guidance for future research directions and practical implementations. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

Maximizing Learning Outcomes while Minimizing: Costs: A Cost-Effective Approach to M-Learning

International Journal of Interactive Mobile Technologies

引用

International Journal of Interactive Mobile Technologies 2024年第11期18卷 4-14页

作者： Suresh Kumar, B. Hemawathi, M.S. Dhanasekaran, P. Kaliappan, Vishnu Kumar Sivaganesan, D. Department of Electronics and Communication Engineering The Kavery Engineering College Tamil Nadu Salem India Department of Computer Science and Engineering KPR Institute of Engineering and Technology Tamil Nadu Coimbatore India Department of Computer Science and Engineering PSG Institute of Technology and Applied Research Tamil Nadu Coimbatore India

In the realm of education, the pursuit of effective learning outcomes often faces the challenge of limited resources. This paper explores the intersection of maximizing learning outcomes and minimizing costs through a cost-effective approach to mobile learning (m-learning). Recognizing the ubiquitous presence of mobile devices and their potential to revolutionize learning, this study explores strategies that leverage the capabilities of mobile technology while being mindful of budgetary constraints. By leveraging the flexibility and accessibil- ity offered by m-learning platforms, educators can design engaging and interactive learning experiences that cater to diverse learners needs. The paper investigates various cost-effective methodologies, including utilizing open-source software, repurposing existing resources, and adopting collaborative learning environments. Furthermore, it examines the role of scalable technologies and adaptive learning algorithms in optimizing resource allocation and per- sonalizing learning experiences. Through case studies and empirical analysis, this research illustrates how institutions and educators can achieve significant cost savings without com- promising the quality of educational delivery. Ultimately, the findings highlight the transfor- mative potential of a cost-effective approach to m-learning in enhancing learning outcomes and widening access to education. © 2024 by the authors of this article.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Modified 2:1 Multiplexer-Based Low Power Ternary ALU for IoT Applications

引用

Journal of engineering science and Technology Review 2024年第5期17卷 104-109页

作者： Allwin Devaraj, S. Magdalin Mary, D. Kannan, P. Esakki Rajavel, S. Thangaraj, Cynthia Anbuselvi Gurumoorthy, K.B. William, Blanie Scrimshaw Department of Electronics and Communication Engineering Francis Xavier Engineering College Tamilnadu Tirunelveli627003 India Department of Electrical and Electronics Engineering Sri Krishna College of Technology Tamilnadu Coimbatore641042 India Department of Electronics and Communication Engineering Karpagam Academy of Higher Education Tamilnadu Coimbatore641021 India Department of Electronics and Communication Engineering SEA College of Engineering and Technology Karnataka Bengaluru560049 India Department of Electronics and Communication Engineering KPR Institute of Engineering and Technology Tamilnadu Coimbatore641407 India Department of Computer Science and Engineering Rohini College of Engineering and Technology Tamilnadu Kanyakumari629401 India

The ternary logic has a benefit over the binary logic which provides a secured solution to achieve a trade-off between the area and power of the design. However, from the structure of the ternary Aritmetic Logic Unit (ALU), it is clear that its architecture increases the area, propagation delay, and power consumption. To overcome this drawback, a loopback algorithm is proposed to achieve low power and high throughput Internet of Things (IoT) processors. The loopback algorithm reduces the number of processing stages in multipliers and adders which can significantly reduce area and power dissipation. The proposed 2:1 multiplexer-based approach reduces the need for a decoder and results in low power consumption. The proposed design will be implemented in Xilinx ISE 13.0 and simulation will be done in Modelsim. The modified Ternary ALU (TALU) performs finer than the previous TALU method. The number of registers used in this architecture is reduced by up to 25% than the existing system therefore there is a reduction in power dissipation. © (2024), (International Hellenic University - School of science). All rights reserved.

关键词： Logic design

来源：评论

学校读者我要写书评

暂无评论

Deep learning based video surveillance for predicting vehicle density in real time scenario

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第4期14卷 4371-4383页

作者： Priyanka, G. Senthil Kumar, J. Veena, S.T. Department of Computer Science and Engineering Mepco Schlenk Engineering College Sivakasi India Department of Electronics and Communication Engineering Mepco Schlenk Engineering College Sivakasi India

Video surveillance systems are often used for traffic monitoring and to characterize traffic load. However, most of the surveillance videos are low frame rated and extracting the right motion feature from them is a challenging task. Deep learning is a growing technology that extracts each object’s feature without human intervention. This paper proposes two different Convolutional Neural Networks (CNN) named Multi Scale Convolutional Neural networks (MSCNN) and Single Scale Convolutional Neural networks (SSCNN) to extract the features and prediction of traffic density. The experimental evaluation of the proposed models was carried out on the UCSD traffic dataset. Our method exploits the benefits of the redundant nature of traffic scenes that are captured from a top-down perspective showing vehicles on a road encircled by background objects. Though a variety of objects is present in each frame, our proposed method attempts to acquire information only about specific object features which helps to predict traffic density better in real-time scenarios. Based on the extracted features, each traffic scene is categorized into three different classes low, medium, and high. Also, the performance of the proposed architecture exhibits 12% more accuracy compared with the existing deep neural networks. Furthermore, in comparison with the MAE score, the proposed system is able to reduce the error from 1 to 0.3 than the existing models. In addition, the training and testing phases of the proposed approach consume considerably less time and optimized usage of CPU resources. © 2023, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

A Credit Card Fraud Model Prediction Method Based on Penalty Factor Optimization AWTadaboost

引用

computers, Materials & Continua 2023年第3期74卷 5951-5965页

作者： Wang Ning Siliang Chen Fu Qiang Haitao Tang Shen Jie College of Computer and Communication Hunan Institute of EngineeringXiangtan411104China College of Computational Science and Electronics Hunan Institute of EngineeringXiangtan411104China

With the popularity of online payment, how to perform creditcard fraud detection more accurately has also become a hot issue. And withthe emergence of the adaptive boosting algorithm (Adaboost), credit cardfraud detection has started to use this method in large numbers, but thetraditional Adaboost is prone to overfitting in the presence of noisy ***, in order to alleviate this phenomenon, this paper proposes a newidea: using the number of consecutive sample misclassifications to determinethe noisy samples, while constructing a penalty factor to reconstruct thesample weight assignment. Firstly, the theoretical analysis shows that thetraditional Adaboost method is overfitting in a noisy training set, which leadsto the degradation of classification accuracy. To this end, the penalty factorconstructed by the number of consecutive misclassifications of samples isused to reconstruct the sample weight assignment to prevent the classifierfrom over-focusing on noisy samples, and its reasonableness is ***, by comparing the penalty strength of the three different penalty factorsproposed in this paper, a more reasonable penalty factor is ***, in order to make the constructed model more in line with theactual requirements on training time consumption, the Adaboost algorithmwith adaptive weight trimming (AWTAdaboost) is used in this paper, so thepenalty factor-based AWTAdaboost (PF_AWTAdaboost) is finally ***, PF_AWTAdaboost is experimentally validated against other traditionalmachine learning algorithms on credit card fraud datasets and otherdatasets. The results show that the PF_AWTAdaboost method has betterperformance, including detection accuracy, model recall and robustness, thanother methods on the credit card fraud dataset. And the PF_AWTAdaboostmethod also shows excellent generalization performance on other *** the experimental results, it is shown that the PF_AWTAdaboost algorithmhas better classification

关键词： Credit card fraud noisy samples penalty factors AWTadaboost algorithm

来源：评论

学校读者我要写书评

暂无评论

Attribute preserving recommendation system based on graph attention mechanism

引用

Journal of Intelligent and Fuzzy Systems 2023年第6期44卷 9419-9430页

作者： Sangeetha, M. Thiagarajan, Meera Devi Department of Computer Science and Engineering Kongu Engineering College Tamilnadu Erode India Department of Electronics and Communication Engineering Kongu Engineering College Tamilnadu Erode India

A recommendation System (RS) is an emerging technology to figure out the user's interests and intentions. As the amount of data increases exponentially, it is hard to analyze the user intentions and trigger the recommendation accordingly. In this research work, a novel recommendation system called the Deep Knowledge Graph based Attribute Preserving Recommendation (DKG-APR) is presented to analyze massive data and provide personalized recommendations to users. The Deep Knowledge Graph for Recommendation System (DKG-RS) uses Deep Convolutional Neural Network (DCNN) and attention mechanism to explicitly model high-order connections in knowledge graphs. According to empirical findings, Knowledge Graph Attention Network (KGAT) performs better than other state-of-the-art recommendation techniques like RippleNet and Neural FM. Additional research demonstrates the effectiveness of embedding propagation for high-order relation modeling and the advantages of the attention mechanism for *** results also show that user information is crucial in the recommendation system, as seen from the optimal node-drop-out ratio of 0.2, which led to the best recall value of 0.2 for all datasets. © 2023 - IOS Press. All rights reserved.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

Optimal Deep Convolutional Neural Network with Pose Estimation for Human Activity Recognition

引用

computer Systems science & engineering 2023年第2期44卷 1719-1733页

作者： S.Nandagopal G.Karthy A.Sheryl Oliver M.Subha Department of Computer Science and Engineering Nandha College of TechnologyErode638052TamilnaduIndia Department of Electronics and Communication Engineering Kalasalingam Academy of Research and EducationKrishnankoil626126TamilnaduIndia Department of Computer Science and Engineering St.Joseph's College of EngineeringChennai600119TamilnaduIndia Department of Electronics and Communication Engineering University College of Engineering NagercoilNagercoil629004TamilnaduIndia

Human Action Recognition(HAR)and pose estimation from videos have gained signiﬁcant attention among research communities due to its applica-tion in several areas namely intelligent surveillance,human robot interaction,robot vision,*** considerable improvements have been made in recent days,design of an effective and accurate action recognition model is yet a difﬁcult process owing to the existence of different obstacles such as variations in camera angle,occlusion,background,movement speed,and so *** the literature,it is observed that hard to deal with the temporal dimension in the action recognition *** neural network(CNN)models could be used widely to solve *** this motivation,this study designs a novel key point extraction with deep convolutional neural networks based pose estimation(KPE-DCNN)model for activity *** KPE-DCNN technique initially converts the input video into a sequence of frames followed by a three stage process namely key point extraction,hyperparameter tuning,and pose *** the keypoint extraction process an OpenPose model is designed to compute the accurate key-points in the human ***,an optimal DCNN model is developed to classify the human activities label based on the extracted key *** improving the training process of the DCNN technique,RMSProp optimizer is used to optimally adjust the hyperparameters such as learning rate,batch size,and epoch *** experimental results tested using benchmark dataset like UCF sports dataset showed that KPE-DCNN technique is able to achieve good results compared with benchmark algorithms like CNN,DBN,SVM,STAL,T-CNN and so on.

关键词： Human activity recognition pose estimation key point extraction classiﬁcation deep learning RMSProp

来源：评论

学校读者我要写书评

暂无评论

Sawtooth-enhanced bend sensor for gesture recognition

引用

science China(Technological sciences) 2024年第6期67卷 1727-1736页

作者： BAI YanRu ZHANG ZiHang WANG HaoYu GUO Rui LI XiSheng School of Automation and Electrical Engineering University of Science and Technology BeijingBeijing100083China School of Advanced Engineering University of Science and Technology BeijingBeijing100083China School of Precision Instrument and Opto-Electronics Engineering Tianjin UniversityTianjin300072China School of Computer and Communication Engineering University of Science and Technology BeijingBeijing100083China

Gesture recognition has diverse application prospects in the field of human-computer ***,gesture recognition devices based on strain sensors have achieved remarkable results,among which liquid metal materials have considerable advantages due to their high tensile strength and *** improve the detection sensitivity of liquid metal strain sensors,a sawtooth-enhanced bending sensor is proposed in this *** with the results from previous studies,the bending sensor shows enhanced resistance *** addition,combined with machine learning algorithms,a gesture recognition glove based on the sawtooth-enhanced bending sensor is also fabricated in this study,and various gestures are accurately *** the fields of human-computer interaction,wearable sensing,and medical health,the sawtooth-enhanced bending sensor shows great potential and can have wide application prospects.

关键词： liquid metal bending sensor gesture recognition machine learning

来源：评论

学校读者我要写书评

暂无评论

Achieving Secured Medical Network (SMN) through Stateless Mechanism and SkeyM in Medical-Internet of Things (M-IoT)

引用

Journal of engineering and Applied science 2024年第1期71卷 1-22页

作者： Nithya, S. Palanisamy, SatheeshKumar Nivethitha, T. Department of Computer Science and Engineering Kumaraguru College of Technology Tamil Nadu Coimbatore641049 India Department of Electronics and Communication Engineering BMS Institute of Technology and Management Karnataka Bengaluru560064 India Department of Electronics and Communication Engineering Sri Eshwar College of Engineering Tamil Nadu Coimbatore641202 India

Medical Internet of Things (M-IoT) synchronizes medical devices in a network to provide smart healthcare monitoring to doctors and to provide an interactive model for patients. This embedded networked system gained lots of importance in the last few years. Now almost every hospital adopts M-IoT. Though it had a lot of challenges in the initial stages of implementation such as feasibility, accuracy, and autonomy, now it has bridged the flaws with the help of several researchers in this field. But still, M-IoT lags behind in security and privacy aspects due to which attackers can intrude on the network and exploit patients’ health data. By examining the various intensive security threats imposed on M-IoT a unique way of handling and transferring data over the network is proposed in this paper. From our research findings, this would resolve the data security issues in M-IoT and commence the next advancements. © The Author(s) 2024.

关键词： Internet of things

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：