检索结果-内蒙古大学图书馆

CVTD: A Robust Car-Mounted Video Text Detector

computers, Materials & Continua 2024年第2期78卷 1821-1842页

作者： Di Zhou Jianxun Zhang Chao Li Yifan Guo Bowen Li Department of Computer Science and Engineering Chongqing University of TechnologyChongqingChina College of Information and Engineering Jingdezhen Ceramic UniversityJingdezhenChina

Text perception is crucial for understanding the semantics of outdoor scenes,making it a key requirement for building intelligent systems for driver assistance or autonomous *** information in car-mounted videos can assist drivers in making ***,Car-mounted video text images pose challenges such as complex backgrounds,small fonts,and the need for real-time *** proposed a robust Car-mounted Video Text Detector(CVTD).It is a lightweight text detection model based on ResNet18 for feature extraction,capable of detecting text in arbitrary *** model efficiently extracted global text positions through the Coordinate Attention Threshold Activation(CATA)and enhanced the representation capability through stacking two Feature Pyramid Enhancement Fusion Modules(FPEFM),strengthening feature representation,and integrating text local features and global position information,reinforcing the representation capability of the CVTD *** enhanced feature maps,when acted upon by Text Activation Maps(TAM),effectively distinguished text foreground from non-text ***,we collected and annotated a dataset containing 2200 images of Car-mounted Video Text(CVT)under various road conditions for training and evaluating our model’s *** further tested our model on four other challenging public natural scene text detection benchmark datasets,demonstrating its strong generalization ability and real-time detection *** model holds potential for practical applications in real-world scenarios.

关键词： Deep learning text detection Car-mounted video text detector intelligent driving assistance arbitrary shape text detector

来源：评论

学校读者我要写书评

暂无评论

Formal verification for security and attacks in IoT physical layer

引用

Journal of Reliable Intelligent Environments 2024年第1期10卷 73-91页

作者： Toman, Zinah Hussein Hamel, Lazhar Toman, Sarah Hussein Graiet, Mohamed Valadares, Dalton Cézane Gomes Department of Computer Science FSM Al-Monastir University Monastir Tunisia Department of Computer Science Faculty of Computer Science and Information Technology Al-Qadisiyah University Al Diwaniyah Iraq Department of Computer Science ISIMM Al-Monastir University Monastir Tunisia IMT Atlantique LS2N Nantes France Federal Institute of Pernambuco/Federal University of Campina Grande Campina Grande Brazil

IoT devices are more important than ever. In a connected world, IoT devices have many uses. They are no longer merely used at work;they are part of our everyday lives. Security concerns arise if the devices generate, collect, or process sensitive data. Physical layer security controls are the cornerstone once the risk for humans increases when physical security fails. To achieve security in IoT devices, preventing is better than detecting. Formal verification is an important and valuable tool for detecting possible vulnerabilities and ensuring data security. Thus, this paper proposes an Event-B proof-based formal model of IoT physical layer security and attacks from the requirements analysis level to the goal level. Our model is built incrementally using a refining method during design and verification. We present a three-level formal approach: first, the construction of the IoT physical layer;then, we check for IoT physical layer vulnerabilities by processing the lack of some characteristics that cause these vulnerabilities, such as speed, typical bandwidth, and power consumption;lastly, we detect physical layer attacks like jamming and MAC spoofing, which helps to build security proofs. Our approach uses an electrocardiogram (ECG) IoT system as a case study, and as an additional case study to back up the proposed method’s generalizability, we used a fire alarm system. Also, we use the proof obligations and the ProB animator in the Rodin model checking tool to check and validate our approach. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning and Powerful Computational Framework for Brain Cancer MRI Image Recognition

引用

Journal of The Institution of Engineers (India): Series B 2024年第1期105卷 1-18页

作者： Kumar, Ankit Shukla, Santosh Kumar Prakash, Navin Yadav, Rakesh Kumar Department of Information Technology Babu Banarasi Das Institute of Technology and Management Lucknow India Department of Computer Science and Engineering Babu Banarasi Das Engineering College Lucknow India Department of Computer Science and Engineering MSOET Maharishi University of Information Technology Lucknow India

Brain tumors are one of the deadliest diseases and require quick and accurate methods of detection. Finding the optimum image for research goals is the first step in optimizing MRI images for pre- and post-processing. As a result, the mean grey-level approach was used to segment the MRI images using a threshold. In the second stage of statistical feature analysis, the picture features were extracted using the spatial gray-level dependency matrix using Hara lick's feature equations. As a consequence, the best features were picked and the tumor was placed in the appropriate location. In the third stage, an automated tool for classifying the photographs under assessment as having a tumor or not was created utilizing supervised learning and an artificially intelligent methodology. An efficient test of the network's performance yielded 97% of the desired results. © 2023, The Institution of Engineers (India).

关键词： Tumors

来源：评论

学校读者我要写书评

暂无评论

An End-to-End Transformer-Based Automatic Speech Recognition for Qur’an Reciters

引用

computers, Materials & Continua 2023年第2期74卷 3471-3487页

作者： Mohammed Hadwan Hamzah A.Alsayadi Salah AL-Hagree Department of Information Technology College of ComputerQassim UniversityBuraydah51452Saudi Arabia Department of Computer Science College of Applied SciencesTaiz UniversityTaiz6803Yemen Computer Science Department Faculty of Computer and Information SciencesAin Shams UniversityCairo11566Egypt Computer Science Department Faculty of SciencesIbb UniversityYemen Department of Computer Sciences&Information Ibb UniversityYemen

The attention-based encoder-decoder technique,known as the trans-former,is used to enhance the performance of end-to-end automatic speech recognition(ASR).This research focuses on applying ASR end-toend transformer-based models for the Arabic language,as the researchers’community pays little attention to *** Muslims Holy Qur’an book is written using Arabic diacritized *** this paper,an end-to-end transformer model to building a robust Qur’an *** is *** acoustic model was built using the transformer-based model as deep learning by the PyTorch framework.A multi-head attention mechanism is utilized to represent the encoder and decoder in the acoustic *** filter bank is used for feature *** build a language model(LM),the Recurrent Neural Network(RNN)and Long short-term memory(LSTM)were used to train an n-gram word-based *** a part of this research,a new dataset of Qur’an verses and their associated transcripts were collected and processed for training and evaluating the proposed model,consisting of 10 h *** recitations performed by 60 *** experimental results showed that the proposed end-to-end transformer-based model achieved a significant low character error rate(CER)of 1.98%and a word error rate(WER)of 6.16%.We have achieved state-of-the-art end-to-end transformer-based recognition for Qur’an reciters.

关键词： Attention-based encoder-decoder recurrent neural network long short-term memory qur’an reciters recognition diacritized arabic text

来源：评论

学校读者我要写书评

暂无评论

A Secure Transmission of Encrypted Medical Data Based on Virtual Instruments 6th

A Secure Transmission of Encrypted Medical Data Based on Vir...

引用

6th International Conference on Innovative Computing and Communication, ICICC 2023

作者： Abdulbaqi, Azmi Shawkat Department of Computer Science College of Computer Science and Information Technology University of Anbar Ramadi Iraq

ISBN: (纸本)9789819940707

Cardiovascular health is becoming more common over the world. Electrocardiography (ECG) is a test that utilizes an electrical signal in the heart to diagnose and monitor cardiac problems. Utilizing the myDAQ data acquisition device and LabVIEW, identifying cardiac illness in real-time is now not only possible but also simple. As a result, this article presents a system that can capture ECG signals in real-time, detect heart problems, and reveal whether a waveform is in range or not utilizing light-emitting diodes (LEDs). LabVIEW software was utilized to create and evaluate the ECG signal of a patient and to calculate associated parameters for the prototype home ECG system described in this article. These parameters include–the Rate of Heart, intervals-of-PR, intervals-of-QT, intervals-of-ST, waves-of-P, and waves-of-T durations. The distant station receives the recorded ECG signal through the Internet, as well as its related factors. Tests were done on the system utilizing ECG’s simulator, web-based-ECG, and actual ECG acquisition. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2024.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Predicting Student Behavior Using a Neutrosophic Deep Learning Model

引用

Neutrosophic Sets and Systems 2025年 76卷 288-310页

作者： Shitaya, Ahmed Mohamed Wahed, Mohamed El Syed Abd El Khalek, Saied Helemy Ismail, Amr Shams, Mahmoud Y. Salama, A.A. Department of Mathematics and Computer Science Faculty of Science Port Said University Egypt Department of Computer Science Faculty of computer and information sciences Suez Canal University Egypt Department. of Communication and Computer Engineering Air Defense University Egypt Department of Cybersecurity College of Engineering and Information Technology Buraydah Private Colleges Buraydah51418 Saudi Arabia Faculty of Artificial Intelligence Kafrelsheikh University Kafrelsheikh33516 Egypt

We developed an information system using an object-oriented programming language and a distributed database (DDB) consisting of multiple interconnected databases across a computer network, managed by a distributed database management system (DDBMS) for easy access. An intelligent system was designed to assess the difficulty level of preliminary exams and select top-performing advanced students using a Neutrosophic Deep Learning Model. The dataset was randomly split into training (80%) and testing (20%) sets, and the model, trained with the Adam optimizer at a 0.001 learning rate over 50 epochs, incorporated early stopping based on validation loss. This system, implemented at a traditional Egyptian university, achieved a 95% accuracy in predicting student dropout. Student behavior, influenced by personal, environmental, and academic factors, is often evaluated subjectively, leading to inconsistent results. Traditional machine learning approaches struggle with the inherent uncertainty in behavioral data. To address this, we combined neutrosophic theory—a mathematical framework that accounts for truth, falsity, and indeterminacy—with deep learning, which excels at learning complex data relationships, to predict student outcomes such as dropout rates. Evaluating the model on student data, including attendance and grades, showed superior accuracy, achieving a determination coefficient of 0.95, demonstrating the approach's potential for identifying at-risk students and enabling targeted interventions. © 2025, university of New Mexico. All rights reserved.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Novel High-Efficiency Transaction Verification Scheme for Blockchain Systems

引用

computer Modeling in Engineering & sciences 2024年第5期139卷 1613-1633页

作者： Jingyu Zhang Pian Zhou Jin Wang Osama Alfarraj Saurabh Singh Min Zhu School of Computer&Communication Engineering Changsha University of Science&TechnologyChangsha410015China Science and Technology on Information Systems Engineering Laboratory School of Systems EngineeringNational University of Defense TechnologyChangsha410073China Computer Science Department King Saud UniversityRiyadh11437Saudi Arabia Department of Artificial Intelligence and Big Data Woosong UniversityDaejeon34606South Korea College of Information Science and Technology Zhejiang Shuren UniversityHangzhou310015China

Blockchain can realize the reliable storage of a large amount of data that is chronologically related and verifiable within the *** technology has been widely used and has developed rapidly in big data systems across various *** increasing number of users are participating in application systems that use blockchain as their underlying *** the number of transactions and the capital involved in blockchain grow,ensuring information security becomes *** the verification of transactional information security and privacy has emerged as a critical ***-based verification methods can effectively eliminate the need for centralized third-party ***,the efficiency of nodes in storing and verifying blockchain data faces unprecedented *** address this issue,this paper introduces an efficient verification scheme for transaction ***,it presents a node evaluation module to estimate the activity level of user nodes participating in transactions,accompanied by a probabilistic analysis for all ***,this paper optimizes the conventional transaction organization form,introduces a heterogeneous Merkle tree storage structure,and designs algorithms for constructing these heterogeneous *** analyses and simulation experiments conclusively demonstrate the superior performance of this *** verifying the same number of transactions,the heterogeneous Merkle tree transmits less data and is more efficient than traditional *** findings indicate that the heterogeneous Merkle tree structure is suitable for various blockchain applications,including the Internet of *** scheme can markedly enhance the efficiency of information verification and bolster the security of distributed systems.

关键词： Blockchain architecture transaction verification information security heterogeneous Merkle tree distributed systems

来源：评论

学校读者我要写书评

暂无评论

An Asynchronous Data Transmission Policy for Task Offloading in Edge-Computing Enabled Ultra-Dense IoT

引用

computers, Materials & Continua 2024年第12期81卷 4465-4483页

作者： Dayong Wang Kamalrulnizam Bin Abu Bakar Babangida Isyaku Liping Lei Department of Computer Science Faculty of ComputingUniversiti Teknologi MalaysiaJohor Bahru81310JohorMalaysia Department of Computer Science Faculty of Information Communication TechnologySule Lamido UniversityKafin Hausa741103JigawaNigeria School of Political Science and Public Administration East China University of Political Science and LawShanghai201620China

In recent years,task offloading and its scheduling optimization have emerged as widely discussed and signif-icant *** multi-objective optimization problems inherent in this domain,particularly those related to resource allocation,have been extensively ***,existing studies predominantly focus on matching suitable computational resources for task offloading requests,often overlooking the optimization of the task data transmission *** inefficiency in data transmission leads to delays in the arrival of task data at computational nodes within the edge network,resulting in increased service times due to elevated network transmission latencies and idle computational *** address this gap,we propose an Asynchronous Data Transmission Policy(ADTP)for optimizing data transmission for task offloading in edge-computing enabled ultra-dense *** dynamically generates data transmission scheduling strategies by jointly considering task offloading decisions and the fluctuating operational states of edge computing-enabled IoT *** contrast to existing methods,the Deep Deterministic Policy Gradient(DDPG)based task data transmission scheduling module works asynchronously with the Deep Q-Network(DQN)based Virtual Machine(VM)selection module in *** significantly reduces the computational space required for the scheduling *** continuous dynamic adjustment of data transmission bandwidth ensures timely delivery of task data and optimal utilization of network bandwidth *** reduces the task completion time and minimizes the failure rate caused by ***,the VM selection module only performs the next inference step when a new task arrives or when a task finishes its *** a result,the wastage of computational resources is further *** simulation results indicate that the proposed ADTP reduced average data transmission delay and service time by 7.11%and 8.09%,***,the tas

关键词： Bandwidth allocation edge computing internet of things task offloading reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Human Age and Gender Prediction from Facial Images Using Deep Learning Methods 15

Human Age and Gender Prediction from Facial Images Using Dee...

引用

15th International Conference on Ambient Systems, Networks and Technologies Networks, ANT 2024 / The 7th International Conference on Emerging Data and Industry 4.0, EDI40 2024

作者： Dey, Puja Mahmud, Tanjim Chowdhury, Mohammad Sanaullah Hossain, Mohammad Shahadat Andersson, Karl Department of Computer Science and Engineering University of Chittagong Chittagong Bangladesh Department of Computer Science and Engineering Rangamati Science and Technology University Bangladesh Lulea University of Technology Skelleftea Sweden

Human age and gender prediction from facial images has garnered significant attention due to its importance in various applications. Traditional models struggle with large-scale variations in unfiltered images. Convolutional Neural Networks (CNNs) have emerged as effective tools for facial analysis due to their robust performance. This paper presents a novel CNN approach for robust age and gender classification using unconstrained real-world images. The CNN architecture includes convolution, pooling, and fully connected layers for feature extraction, dimension reduction, and mapping to output classes. Adience and UTKFace datasets were utilized, with the best training and testing accuracies achieved using an 80% training and 20% testing data split. Robust image pre-processing and data augmentation techniques were applied to handle dataset variations. The proposed approach outperformed existing methods, achieving age prediction accuracies of 86.42% and 81.96%, and gender prediction accuracies of 97.65% and 96.32% on the Adience and UTKFace datasets, respectively. © 2024 Elsevier B.V.. All rights reserved.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Stacking Ensemble Framework for Hate Speech Detection in Bangla Videos

Stacking Ensemble Framework for Hate Speech Detection in Ban...

引用

2024 IEEE International Conference on Computing, Applications and Systems, COMPAS 2024

作者： Mim, Shamrin Jahan Mahmud, Tanjim Ali, Md Hasan Aziz, Mohammad Tarek Rangamati Science and Technology University Department of Computer Science and Engineering Rangamati4500 Bangladesh Chittagong University of Engineering and Technology Department of Computer Science and Engineering Chittagong Bangladesh

ISBN: (纸本)9798331529765

The proliferation of internet usage and social media platforms has significantly enhanced the ability of individuals to express their opinions on various topics. However, this freedom of expression sometimes morphs into a vehicle for disseminating hate speech, leading to increased incidents of cyberbullying, violations, and conflicts. Particularly on video-sharing websites, which have become a prominent stage for such activities due to their widespread use and the dynamic nature of video content. This study aims to address the issue of hate speech in video content by developing a robust method for detecting hate speech in Bangla language videos. The focus is on the spoken content within these videos, which is a primary vector for the transmission of harmful messages. We constructed a comprehensive dataset by extracting and converting audio from a collection of videos into text. Utilizing this dataset, we applied machine learning techniques and deep learning models to analyze and classify the content. Specifically, our approach involves a stacking ensemble model that combines the strengths of Convolutional Neural Networks (CNN) and Bidirectional Gated Recurrent Units (BiGRU) with the analytical capabilities of customized feature extraction applied to Multinomial Naive Bayes and Random Forest classifiers serving as a meta-model. The proposed stacking ensemble model demonstrates a high level of efficacy, achieving an accuracy rate of 96% in detecting hate speech within the tested video content. This performance indicates a significant advancement over existing methods, underlining the effectiveness of our hybrid, multi-model approach. © 2024 IEEE.

关键词： Convolutional 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：