检索结果-内蒙古大学图书馆

Minimizing Model Size of CNN-Based Vehicle Make Recognition for Frontal Vehicle Images

IEEE Access 2025年 13卷 97409-97420页

作者： Wiput Puisamlee Rathachai Chawuthai 1Department of Electrical Engineering School of Engineering King Mongkut’s Institute of Technology Ladkrabang Bangkok Thailand 2Department of Computer Engineering School of Engineering King Mongkut’s Institute of Technology Ladkrabang Bangkok Thailand

Vehicle Make Model Recognition (VMMR) is commonly used in Intelligent Transportation Systems (ITS), free-flow image-based toll systems, and enforcement systems. These systems must analyze and process vehicle front images for use as evidence. Convolutional Neural Networks (CNN) are widely used for image classification and VMMR problems. Complex model structures and more internal parameters are needed to improve classification accuracy with many classes. Issues included larger models and longer processing times. The goal of this work is to study and create a smaller CNN model that can be used on devices with limited resources, like embedded computers and embedded computer cameras, to figure out what kind of car it is from a front view picture. Real free-flow toll systems were used to train a CNN model that recognized vehicle makes with 99% accuracy. The model is smaller than VGG16, InceptionV3, Yolo11m-cls, and ResNet50 and has over 90% accuracy. It reduced parameters by 69.95% and developed the CTv1 model to achieve an F1 score 2.06% higher than InceptionV3, the best. The model was tested on a Raspberry Pi 3 Model B, processing images in 1 second and using 25 mWh. The compact version of the proposed model also adjusts the Padding and Stride of the Convolutional Layer and reduces the CNN model size using Depth-wise Separable Convolutional and $1\times 1$ Convolutional Dimension Reduction (Bottleneck) methods to test vehicle make recognition accuracy, training time, processing time, and model size.

关键词： Convolution Convolutional neural networks Computational modeling Accuracy Image recognition Filters Feature extraction Kernel Training Residual neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Anxiety Prediction: Leveraging Machine Learning with Functional Design Metrics 21

Enhancing Anxiety Prediction: Leveraging Machine Learning wi...

引用

21st International Conference on Electrical engineering, Computing Science and Automatic Control, CCE 2024

作者： Javed, Abqa Ajmal, Sahar Tajammal, Taliah Aslam, Muhammad Martinez-Enriquez, A.M. Sheikh, Usman Sadiq University of Engineering and Technology Department of Computer Science Lahore Pakistan National University of Computer and Emerging Sciences Department of Software Engineering Pakistan Virtual University of Pakistan Department of Computer Science and Information Technology Lahore Pakistan CINVESTAV-IPN D.F. Department of Computer Science Mexico Medicine Bronglais General Hospital Clinical Fellow in Respiratory SY23 1ER United Kingdom

ISBN: (纸本)9798350377545

Fear, nervousness, and social phobia all contribute to anxiety episodes reaching their peak in seconds. Using data collected from patient questionnaires, this study seeks to anticipate and diagnose anxiety disorders effectively and precisely using machine learning approaches. However, current research offers standardized metrics for assessing machine learning approaches throughout the design process. This work presents a systematic method for efficiently extracting literature from digital sources and emphasizes the need for data engineers to evaluate the efficiency of machine learning approaches to assure forecast quality. The International Organization for Standardization (ISO) 9126 standard is used to assess machine learning approaches for anxiety prediction to address this gap. This evaluation considers parameters such as functionality, accuracy, resilience, throughput, explainability, privacy, and correctness. These defined characteristics serve as the benchmark to measure the efficacy of machine learning approaches in predicting anxiety disorders. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

A Robust DL Approach for Detection of Invasive Ductal Carcinoma in Whole Slide Images using DenseNet169

A Robust DL Approach for Detection of Invasive Ductal Carcin...

引用

2024 International Conference on Communication, computer Sciences and engineering, IC3SE 2024

作者： Menaka, C. Sharma, Sunil Naval, Preeti Brar, Khushmeen Anand, Rohit Alabdeli, Haider Pant, Shivani Department Of Computer Science And Information Technology Bangalore India Vivekananda Global University Department Of Computer Science & Application Jaipur India Maharishi University Of Information Technology Maharishi School Of Engineering & Technology Uttar Pradesh India Chandigarh University Department Of Electronics And Communication Engineering Punjab Mohali India G.B Pant Okhla-1 Campus Department Of Ece New Delhi India The Islamic University Najaf Iraq Uttaranchal University Uttaranchal Institute Of Technology Uttarakhand India

ISBN: (纸本)9798350366846

The present research introduces a deep learning method for automatically recognizing and evaluating tissue areas that correspond to invasive ductal carcinoma (IDC) in whole slide images (WSI) of breast cancer (BCa). Learn-from-data techniques which utilize computational simulations of the learning process are called deep learning approaches. It has been demonstrated that these methodologies outperform conventional methods in resolving the challenging issues in a number of domains, particularly object and speech recognition. The main reason why making an invasive breast cancer evaluation is an exhausting and laborious process is that, it necessitates a pathologist to scan a huge number of benign regions with the aim to determine the spots that are malignant. Effectively characterizing IDC in WSI is a prerequisite for evaluating the severity of tumors and predicting patient prognosis. In order to support diagnosis, the DL approach in this study draws out many convolutional neural networks (CNNs) for perceptible semantic examination of tumor locations. A WSI database of 162 patients with IDC diagnoses is exploited to assess the technique, and 170 slides were chosen for training. The objective of the experimental analysis was to assess the accuracy of the classifier in identifying IDC tissue areas in WSI. A fusion of optimization techniques namely Best First and Evolutionary Search has been employed for feature reduction and improvement. When using DenseNet169, our approach produced the highest quantitative outcomes for the computerized identification of IDC sections in WSI with regard to accuracy 87.46% in comparison with an approach using EfficientNetB0 and ML classifiers for invasive tumor categorization employing Random Forest, BayesNet, SVM, Naïve Bayes and Decision Tree with accuracy of 80.65%. © 2024 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

A comprehensive survey on 6G-security: physical connection and service layers

引用

Discover Internet of Things 2025年第1期5卷 1-30页

作者： Saeed, Mamoon M. Saeed, Rashid A. Hasan, Mohammad Kamrul Ali, Elmustafa Sayed Mazha, Tehseen Shahzad, Tariq Khan, Sunawar Hamam, Habib Sana00967 Yemen Department of Computer Engineering College of Computers and Information Technology Taif University P.O. Box 11099 Taif21944 Saudi Arabia Department of Electronics Engineering College of Electronics Engineering Sudan University of Science and Technology Omdurman Sudan Bangi43600 Malaysia Department of Electrical and Electronic Engineering Faculty of Engineering Red Sea University Port Sudan33312 Sudan School of Computer Science National College of Business Administration and Economics Lahore54000 Pakistan Faculty of Engineering Uni de Moncton MonctonNBE1A3E9 Canada Department of Computer Engineering COMSATS University Islamabad Sahiwal Campus Sahiwal57000 Pakistan Department of Computer Science and Software Engineering International Islamic University Islamabad44000 Pakistan School of Electrical Engineering University of Johannesburg Johannesburg2006 South Africa BP 1989 Av. Grandes Ecoles Libreville Gabon Bridges for Academic Excellence-Spectrum Tunis Tunisia Department of Computer Science School Education Department Government of Punjab Layyah31200 Pakistan Chitkara University Institute of Engineering and Technology Chitkara University Punjab Rajpura140401 India

After adopting 5G technology, businesses and academia have started working on sixth-generation wireless networking (6G) technologies. Mobile communications options are expected to expand in areas where previous generations could not do so. 6G networks are anticipated to be constructed using various diverse technologies. These encompass diverse cutting-edge advancements, such as distributed ledger systems like blockchain, visible light communications (VLC), post-quantum cryptography, edge computing, molecular communication, THz, and other advances. These advances necessitate a reassessment of previous security strategies from a security perspective. In the future, networks must adhere to stricter criteria for authentication, encryption, access control, connectivity, and detection of harmful activities. Ensuring privacy and dependability necessitates the implementation of supplementary security protocols. The essay explores the primary concerns and challenges related to the security of the 6G network. This paper describes the improvements in security in communications from 1G through 6G. This paper divides security in the sixth generation into three layers: physical, connection, and service. Each layer-by-layer discusses the standard technologies and security issues for each technology proposed in each sixth-generation security layer. All proposed solutions for each of the three layers are discussed in Sixth Generation Security. It also reviews all proposed solutions for each layer, indicating the proposed solution and its limitations. © The Author(s) 2025.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

A Zero-Sequence Current Analysis Approach for Rotating Machinery Fault Diagnosis of Induction Motor Drivetrain Based on Sparse Learning

引用

IEEE Transactions on Power Electronics 2025年第7期40卷 9800-9810页

作者： Liu, Zhao-Hua Long, Jun-Jie Wei, Hua-Liang Liu, Kan Zhang, Ying-Jie Li, Xiao-Hua Hunan University of Science and Technology School of Information and Electrical Engineering Xiangtan411201 China The University of Sheffield School of Electrical and Electronic Engineering Department of Automatic Control and Systems Engineering SheffieldS1 3JD United Kingdom Hunan University College of Mechanical and Vehicle Engineering Changsha410082 China Hunan University College of Computer Science and Electronic Engineering Changsha410082 China

Fault diagnosis of rotating machinery driven by induction motors has received increasing attention. Current diagnostic methods, which can be performed on existing inverters or current transformers of three-phase induction machines, have become a more economical and reliable alternative to vibration diagnostic methods. Existing references mainly focus on the stator current analysis of single-phase, but most single-phase current analysis methods utilize only a fraction of the total information accessible in the three-phase system. Field experience shows that zero-sequence current (ZSC) has more distinctive fault characteristics compared to single-phase current, and therefore it is better to use in mechanical fault detection and diagnostic tasks. This article proposes a novel sparse learning-based method for ZSC analysis for induction rotating motor drive fault diagnosis. First, it elaborates and compares the effectiveness of ZSC in revealing fault characteristics compared with single-phase current by analyzing the fault diagnosis mechanism of ZSC. In addition, the method proposes a correlation entropy-enhanced sparse learning method for the problem of high-frequency noise and interference in the ZSC signals, so as to enhance the learning of the features of the noise-containing signals. The diagnostic efficacy of the proposed method has been verified through experiments on two real datasets. © 1986-2012 IEEE.

关键词： Induction motors

来源：评论

学校读者我要写书评

暂无评论

Bangla License Plate Detection and Recognition Approach Based on computer Vision for Authentic Vehicle Identification 3rd

Bangla License Plate Detection and Recognition Approach Base...

引用

3rd International Conference on Trends in Electronics and Health Informatics, TEHI 2023

作者： Dhar, Rajib Kumar Khanom, Fahmida Uddin, Khandaker Mohammad Mohi Imam, Md. Hasan Mostafiz, Rafid Institute of Information Technology Noakhali Science and Technology University Noakhali Bangladesh Department of Mathematics American International University—Bangladesh 408/1 Kuratoli Khilkhet Dhaka1229 Bangladesh Department of Computer Science and Engineering Southeast University Dhaka1208 Bangladesh

ISBN: (纸本)9789819739363

Vehicle license plates (LPs) detection and recognition is a fundamental technique in computer vision that plays a crucial role to extract, localize, and authenticate the information of a vehicle. This paper presents an efficient method for Bangla LPs recognition and vehicle authentication using edge analysis and pattern matching algorithm in real-time nature. This method performs four modules as: preprocessing, license plate identification, character recognition, vehicle authentication. A multi-step preprocessing technique has been applied to extract and understand the potential information in case of any uneven images. To identify edge connectivity and uniformity, Canny edge detector provides more promising in contour analysis and character localization. The segmented characters are converted into texts and verified using the pattern matching algorithm by a predefined database. For this research, a database consisting of 3000 automobile license plate photos captured under different climatic circumstances has been compiled. Additionally, we have curated two additional databases consisting of templates for recognition and registered automobiles for the purpose of authentication. The proposed model can achieve accuracy of 96% in LPs detection, 99% in character recognition, and 100% in vehicle authentication by extracted texts from LPs. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Authentication

来源：评论

学校读者我要写书评

暂无评论

Unveiling vulnerable smart contracts:Toward profiling vulnerable smart contracts using genetic algorithm and generating benchmark dataset

引用

Blockchain(Research and Applications) 2024年第1期5卷 58-84页

作者： Sepideh HajiHosseinKhani Arash Habibi Lashkari Ali Mizani Oskui Computer Science Department of Computer Science and EngineeringYork UniversityTorontoOntario M3J1P3Canada Behavior-Centric Cybersecurity Center(BCCC) School of Information TechnologyYork UniversityTorontoOntario M3J1P3Canada Financial and Crypto Advisory of Switzerland(FiCAS AG) ZugSwitzerland

Smart contracts(SCs)are crucial in maintaining trust within blockchain ***,existing methods for analyzing SC vulnerabilities often lack accuracy and effectiveness,while approaches based on Deep Neural Networks(DNNs)struggle with detecting complex vulnerabilities due to limited data *** paper proposes a novel approach for analyzing SC *** method leverages an advanced form of the Genetic Algorithm(GA)and includes the development of a comprehensive benchmark dataset consisting of 36,670 Solidity source code *** primary objective of our study is to profile vulnerable SCs *** achieve this goal,we have devised an analyzer called SCsVulLyzer based on GAs,designed explicitly for profiling ***,we have carefully curated a new dataset encompassing a wide range of examples,ensuring the practical validation of our ***,we have established three distinct taxonomies that cover SCs,profiling techniques,and feature *** taxonomies provide a systematic classification and analysis of information,improving the efficiency of our *** methodology underwent rigorous testing through experimentation,and the results demonstrated the superior capabilities of our model in detecting *** to traditional and DNN-based approaches,our approach achieved higher precision,recall,and F1-score,which are widely used metrics for evaluating model *** all these metrics,our model showed exceptional *** customization and adaptations we implemented within the GA significantly enhanced its *** approach detects SC vulnerabilities more efficiently and facilitates robust *** promising results highlight the potential of GA-based profiling to improve the detection of SC vulnerabilities,contributing to enhanced security in blockchain networks.

关键词： Smart contracts(SCs) Vulnerability Vulnerable smart contracts Vulnerability profiling Genetic algorithm

来源：评论

学校读者我要写书评

暂无评论

A distinguished-bit tracking knowledge-based query tree for RFID tag identification

引用

computer Communications 2024年 218卷 166-175页

作者： Lin, Chih-Chung Lai, Yuan-Cheng Hailemariam, Zelalem Legese Department of Computer Information and Network Engineering Lunghwa University of Science and Technology No. 300 Sec. 1 Wanshou Rd. Guishan District Taoyuan City 333326 Taiwan Department of Information Management National Taiwan University of Science and Technology No. 43 Sec. 4 Keelung Road Taipei 106 Taiwan

—Many RFID systems own the information of RFID tags in the database, so the reader knows how many tags will be identified and which tags’ IDs might appear. How the RFID anti-collision protocols take advantage of this knowledge is a challenge. Some previous anti-collision protocols, including Query Tree with Shortcutting and Couple Resolution (QTSC) and Bit-tracking Knowledge-based Query Tree (BKQT), solved this task with the knowledge. We propose one novel protocol, Distinguished-bit Tracking Knowledge-based Query Tree (DKQT), which integrates the knowledge with the bit-tracking technique. Utilizing the knowledge and knowing the collided-bit locations, DKQT adopts a distinguishable technique, which can identify the tags that have a distinguished-bit representing that only one tag has a unique “0” or “1” in a bit location of the tag ID, to speed up the tag identification. Innovatively, adopting a distinguishable technique, DKQT forms an n-ary tree where a node has some leaves that are the tags having distinguished-bits. Thus, DKQT using the distinguishable technique can recognize multiple appearing tags in a collision by traversing this n-ary tree. The simulation results show that DKQT can reduce the number of slots by 36.94 % and 15.58 %, and the identification time by 37.47 % and 17.86 %, compared to the previous knowledge-based protocols, QTSC and BKQT, respectively. © 2024 Elsevier B.V.

关键词： Distinguished-bit Knowledge-based tree RFID anti-collision Tag identification

来源：评论

学校读者我要写书评

暂无评论

When Generative AI Meets Semantic Communication: Optimizing Radio Map Construction and Distribution in Future Mobile Networks

引用

IEEE Network 2025年第3期39卷 47-55页

作者： Zhou, Li Deng, Xinfeng Ning, Zhaolong Zhao, Haitao Wei, Jibo Leung, Victor C. M. National University of Defense Technology College of Electronic Science and Technology Changsha410073 China Chongqing University of Posts and Telecommunications School of Communications and Information Engineering Chongqing400065 China Shenzhen MSU-BIT University AI Research Institute Shenzhen518000 China Shenzhen University College of Computer Science and Software Engineering Shenzhen518000 China The University of British Columbia Department of Electrical and Computer Engineering VancouverBCV6T 1Z4 Canada

With the rapid development of the internet of things and smart cities, the demand for effective spectrum collaboration has grown *** maps play a crucial role in understanding the spatial radio environment, which is essential for wireless applications such as cell planning and radio resource management. However, generating radio maps is a resource-intensive task, as the construction process is complex and the distribution process consumes substantial bandwidth. This article proposes a method that combines semantic communication (SemCom) and generative artificial intelligence (GAI) to optimize the construction and distribution processes of radio maps. By incorporating SemCom, our method transmits only key semantic information during the distribution process, significantly reducing bandwidth usage while ensuring accurate and efficient information transfer. While the diverse generative capabilities of GAI introduce some instability, which could be a limitation in radio map construction, this article achieves precise content decoding through prompts based on multi-modal semantic information, enabling accurate construction and restoration of radio maps. By utilizing SemCom and GAI technologies, the cost of applications of radio maps can be significantly reduced, which is beneficial for stimulating the pace of intelligence evolution in future mobile networks. © 1986-2012 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

Hybrid approach for next basket recommendation system

引用

International Journal of information technology (Singapore) 2023年第3期15卷 1733-1740页

作者： Yannam, V Ramanjaneyulu Kumar, Jitendra Vankayala, Tejaswini Babu, Korra Sathya Department of Computer Science and Engineering National Institute of Technology Rourkela Sector-1 Odisha Rourkela 769008 India Department Indian Institute of Information Technology Design and Manufacturing Kurnool Jagannathagattu Andhra pradesh Kurnool 518007 India

The recommendation system’s task is to predict what a user would like to buy in his/her next shopping period given in shopping history. Many state-of-the-art recommendation systems have relied mainly on user explicit feedback to deliver recommendations to the end-users since implicit feedback does not directly reflect user feedback. A set of items (multiple) that can be recommended for an user is called next basket recommendation. Most of the next basket recommendation systems treat user sequential behavior and common interests differently. These types of systems are only capable to capture local sequential behavior between adjacent baskets. In this study, an effort was made to close the gap between the hybrid approach’s locally and globally collected item sequences. This paper proposes a hybrid recommendation system that deals with both issues, achieving better results. The model consists of an autoencoder to extract rich contextual information from the user implicit feedback and a recurrent neural network to learn a dynamic representation of the user and capture global sequential features between baskets. The proposed approach has been tested on two bench mark datasets and the results indicate that it is more effective than the state-of-the-art recommendation systems. © 2023, The Author(s), under exclusive licence to Bharati Vidyapeeth's Institute of computer Applications and Management.

关键词： Autoencoders Deep learning Hybrid Recurrent neural networks

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：