检索结果-内蒙古大学图书馆

Extracting Facial Features to Detect Deepfake Videos Using Machine Learning

International Journal of Advanced computer science and Applications 2025年第4期16卷 834-842页

作者： Aslam, Ayesha Mir, Jamaluddin Zaman, Gohar Rahman, Atta Salam, Asiya Abdus Ali, Farhan Alhiyafi, Jamal Bakry, Aghiad Gul, Mustafa Jamal Gollapalli, Mohammed Mahmud, Maqsood Department of Computer Science Abbottabad University of Science and Technology Havelian Pakistan Department of Computer Science College of Computer Science and Information Technology Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam31441 Saudi Arabia Department of Computer Information Systems College of Computer Science and Information Technology Imam Abdulrahman Bin Faisal University P.O. Box 1982 Dammam31441 Saudi Arabia College of Electronics and Information Engineering Shenzhen University Shenzhen China Department of Computer Science Kettering University FlintMI United States Department of Business Administration University of York York YO10 5DD Heslington United Kingdom Department of Information Technology & Engineering Sydney Met SydneyNSW2000 Australia School of Computing Ulster University Northern Ireland Belfast United Kingdom

Generative adversarial networks (GANs) have gained popularity for their ability to synthesize images from random inputs in deep learning models. One of the notable applications of this technology is the creation of realistic videos known as deepfakes, which have been misused on social media platforms. The difficulty lies in distinguishing these fake videos from real ones with the naked eye, leading to significant concerns. This study proposes a supervised machine learning approach to effectively differentiate between real and counterfeit videos by detecting visual artifacts. To achieve this, two facial features are extracted: eye blinking and nose position, utilizing landmark detection techniques. Both features were trained on supervised machine learning classifiers and evaluated using the publicly available UADFV and Celeb-DF deepfake datasets. The experiments successfully demonstrate that the proposed method achieves a promising and superior performance, with an area under the curve (AUC) of 97% for deepfake detection in contrast to state-of-the-art methods investigating the same datasets. © (2025), (science and Information Organization). All rights reserved.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Leveraging Hybrid CNN Model for Enhanced Classification on Imbalanced Datasets 1

Leveraging Hybrid CNN Model for Enhanced Classification on I...

引用

1st International Conference on Intelligent Systems and Computational Networks, ICISCN 2025

作者： Murugan, R. Sekar, K. Sariga, G. Nikhilesh, V. Tejashwini, J.S. Dheepa, T. National Institute of Technology Silchar Department of Ece Silchar India Rajalakshmi Engineering College Department of Aiml Chennai India Kongunadu College of Engineering and Technology Department of Computer Science and Engineering Tiruchirappalli India

ISBN: (数字)9798331529246

ISBN: (纸本)9798331529246

This study addresses the challenge of class imbalance in the CIFAR-10 dataset, where certain classes are significantly underrepresented. To tackle this issue, data augmentation and resampling techniques, particularly oversampling, are employed to achieve a balanced class distribution. A Convolutional Neural Network (CNN) is trained using a class-weighted loss function to effectively mitigate residual imbalances. The model's performance is evaluated on test images, with a ResNet model incorporated for comparison. This hybrid approach enables a comprehensive assessment of classification accuracy and robustness. The proposed model achieves accuracy, precision, and recall scores of 0.95, 0.88, and 0.93, respectively. The results demonstrate significant improvements in the prediction accuracy of minority classes, underscoring the effectiveness of the applied methods. This study highlights the critical role of oversampling and class-weighted loss functions in addressing class imbalance challenges, ultimately enhancing the performance of image classification models. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Asynchronous I/O Persistence for In-Memory Database Servers: Leveraging io_uring to Optimize Redis Persistence 17th

Asynchronous I/O Persistence for In-Memory Database Servers:...

引用

17th International Conference on Cloud Computing, CLOUD 2024, Held as Part of the Services Conference Federation, SCF 2024

作者： Chen, Le-Gao Li, Yanzhi Laohakangvalvit, Tipporn Sugaya, Midori Innovative Global Program College of Engineering Shibaura Institute of Technology Tokyo Koto City Japan Department of Computer Science and Engineering College of Engineering Shibaura Institute of Technology Tokyo Koto City Japan

ISBN: (纸本)9783031771521

With constant hardware improvements allowing for increasingly large memory sizes, in-memory database servers have become an attractive option for various cloud applications. Even though in-memory database servers store all data in memory, there is still a need to persist data onto an I/O device for maintenance, updates, or protection against server crashes. In high throughput environments, persistence happens often, significantly hindering server throughput performance. Although popular in-memory database servers such as Redis or Dragonfly DB have adopted asynchronous I/O for socket operations, none have extended this approach for persistence. This results in significantly longer time taken for persistence, leading to prolonged reduced throughput. Given the recent introduction of io_uring, which has demonstrated significant performance improvements over past asynchronous I/O APIs, we have decided to implement it into Redis to evaluate its effectiveness in reducing these performance overheads. Our findings indicate that io_uring can significantly enhance performance in use cases with medium to large payload sizes but may hinder performances in use cases involving small payload sizes due to higher complexity and higher operation cost per datapoint. This method can likely be applied to other servers, such as Dragonfly DB and Memcached, to achieve similar benefits in appropriate scenarios. Additionally, in-memory database server persistence can serve as valuable benchmarks for further io_uring development and optimization. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Benchmarking

来源：评论

学校读者我要写书评

暂无评论

A Lightweight Convolutional Neural Network with Squeeze and Excitation Module for Security Authentication Using Wireless Channel

引用

computers, Materials & Continua 2025年第5期83卷 2025-2040页

作者： Xiaoying Qiu Xiaoyu Ma Guangxu Zhao Jinwei Yu Wenbao Jiang Zhaozhong Guo Maozhi Xu College of Computer Science Beijing Information Science and Technology UniversityBeijing100192China School of Information and Communication Engineering Beijing University of Posts and TelecommunicationsBeijing100876China School of Mathematical Sciences Peking UniversityBeijing100871China

Physical layer authentication(PLA)in the context of the Internet of Things(IoT)has gained significant *** with traditional encryption and blockchain technologies,PLA provides a more computationally efficient alternative to exploiting the properties of the wireless medium *** existing PLA solutions rely on static mechanisms,which are insufficient to address the authentication challenges in fifth generation(5G)and beyond wireless ***,with the massive increase in mobile device access,the communication security of the IoT is vulnerable to spoofing *** overcome the above challenges,this paper proposes a lightweight deep convolutional neural network(CNN)equipped with squeeze and excitation module(SE module)in dynamic wireless environments,namely *** be more specific,a convolution factorization is developed to reduce the complexity of PLA models based on deep ***,an SE module is designed in the deep CNN to enhance useful features andmaximize authentication *** with the existing solutions,the proposed SE-ConvNet enabled PLA scheme performs excellently in mobile and time-varying wireless environments while maintaining lower computational complexity.

关键词： Physical layer authentication blockchain squeeze and excitation module computational cost mobile scenario

来源：评论

学校读者我要写书评

暂无评论

Live Audio Recognition and Text Conversion System 2

Live Audio Recognition and Text Conversion System

引用

2nd International Conference on Computational Intelligence, Communication technology and Networking, CICTN 2025

作者： Sharma, Pratyaksha Gupta, Sumita Bhavya, Boddu Asmitha Sharma, Khushi Dhir, Saru Gambhir, Sapna Department of Computer Science & Engineering Amity School of Engineering & Technology Amity University UP India Department of Information Technology Amity School of Engineering & Technology Amity University UP India Dept. of Computer Science College of Engineering and Computing George Mason University United States

ISBN: (纸本)9798331530389

With the advancement of technologies, different methods are currently being used for converting spoken language into text. These systems offer a hands-free alternative to traditional input methods, especially for individuals with physical disabilities or those requiring efficient documentation of spoken content. The goal of this study is to review recent advancements in this field and highlight gaps in current research, with a focus on accuracy, language adaptability, ease of use, and the integration of various technologies. We also suggest future directions for enhancing and developing this technology further. © 2025 IEEE.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Fast Detection and Traceback-Based Mitigation of Interest Flooding Attack

引用

SN computer science 2025年第2期6卷 1-15页

作者： Kumar, Naveen Singh, Ashutosh Kumar Srivastava, Shashank Department of Computer Science and Engineering Sardar Vallabhbhai National Institute of Technology Gujarat Surat 395007 India Department of Computer Science Allahabad Degree College University of Allahabad Uttar Pradesh Prayagraj 211003 India Computer Science and Engineering Motilal Nehru National Institute of Technology Allahabad Uttar Pradesh Prayagraj 211004 India

The Pending Interest Table (PIT) in Named Data Networking (NDN) plays a crucial role by storing state information of requests within the router, enabling efficient data packet routing back to the requester. However, this mechanism is vulnerable to Interest Flooding Attacks (IFA), where an attacker sends a large number of malicious requests to overwhelm the PIT, disrupting network performance. Previous research primarily focused on offline detection of IFA using selected features and machine learning techniques. In this work, we build on these findings by deploying a trained Artificial Neural Network (ANN) classifier on each NDN router for real-time, online detection of IFA. Additionally, we introduce a novel traceback-based mitigation strategy activated upon detection, significantly improving the network’s resilience against such attacks. Our proposed method demonstrates superior performance in terms of satisfaction ratio and throughput for legitimate consumers compared to existing approach. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2025.

关键词： ANN Feature selection Interest Flooding Attack Named Data Networking NDN Traceback

来源：评论

学校读者我要写书评

暂无评论

Implementation of YOLOv9-Based Real-Time Object Detection for Robotic Finger Applications 5

Implementation of YOLOv9-Based Real-Time Object Detection fo...

引用

5th IEEE International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies, ICAECT 2025

作者： Duraisamy, Premkumar Deepika, A. Dhanushiya, S. Karthik, S. Kpr Institute of Engineering and Technology Department of Computer Science and Engineering Coimbatore India Hindusthan College of Engineering & Technology Department of Information Technology Coimbatore India

ISBN: (纸本)9798350356236

The project investigates the integration of YOLO v9, the latest iteration of the 'You Only Look Once' model, into robotic fingers to enhance their precision, responsiveness, and versatility in object handling. YOLO v9's superior object detection capabilities - especially its real-time processing speed and high accuracy - make it an ideal candidate for advancing robotic manipulation. By training YOLO v9 on an extensive dataset encompassing diverse object shapes, sizes, and textures, this project enables robotic fingers to recognize and respond to various physical characteristics with greater nuance. Through this approach, robotic systems gain an adaptive edge, adjusting their grip and interaction based on specific object features, thus expanding their utility in precision-demanding fields such as manufacturing, healthcare, and logistics. The research not only explores technical integration but also assesses the impact of this enhanced object recognition on robotics' overall reliability and efficiency in handling real-world tasks. The findings could contribute to the development of more intelligent, responsive, and adaptable robotic systems, redefining automation standards in critical industries. © 2025 IEEE.

关键词： Intelligent robots

来源：评论

学校读者我要写书评

暂无评论

Monkeypox Disease Classification from Skin Lesion Images using Deep Convolution Neural Network 3

Monkeypox Disease Classification from Skin Lesion Images usi...

引用

3rd International Conference on Intelligent Systems, Advanced Computing, and Communication, ISACC 2025

作者： Thangaraj, Rajasekaran Vadivel, Balasubramaniam Kumar, S. Manoj Sadesh, S. Karunakaran, P. Prakash, P. Nandha Engineering College Department of Computer Science and Engineering Erode India Sri Ramakrishna Institute of Technology Department of Computer Science and Engineering Coimbatore India KPR Institute of Engineering and Technology Department of Computer Science and Engineering Coimbatore India Velalar College of Engineering and Technology Department of Artificial Intelligence and Data Science Erode India Nandha Engineering College Department of Artificial Intelligence and Data Science Erode India Velalar College of Engineering and Technology Department of Information Technology Erode India

ISBN: (纸本)9798331523893

Recently monkeypox outbreak has raised concerns due to its increasing number of cases and diverse dermatological symptoms in 2024, which can complicate early diagnosis due to similarities with other viral infections such as measles and chickenpox. To enhance diagnostic accuracy, transfer learning models and artificial intelligence (AI) have been explored as effective tools. Pre-trained models such as VGG16, VGG19, ResNet50V2, and MobileNetV2 have been employed for monkeypox detection, each with distinct trade-offs in terms of computational efficiency and diagnostic precision. MobileNetV2 offers superior efficiency, but this comes at the cost of reduced accuracy. VGG models provide higher accuracy but require much more computational *** these models, DenseNet121 stands out by achieving 99% accuracy while requiring 37% less computational power, making it the model optimized choice for efficient and accurate monkeypox classification. © 2025 IEEE.

关键词： Transfer learning

来源：评论

学校读者我要写书评

暂无评论

Fusion of Multimodal Audio Data for Enhanced Speaker Identification Using Kolmogorov-Arnold Networks

引用

IEEE Access 2025年 13卷 87640-87653页

作者： Tamotia, Aryaman Karmokar, Dhruv Shantu Komal, Rethi Khadar Nawas, K. Shahina, A. Nayeemulla Khan, A. Vellore Institute of Technology School of Computer Science and Engineering Chennai600127 India Sri Sivasubramaniya Nadar College of Engineering Department of Information Technology Tamil Nadu Kalavakkam603110 India

Speaker identification using audio data is quite challenging because of inherent differences between people, ambient noise, and variable recording conditions. Although the classical deep learning methods are effective, they have rather high computational cost, which leads to usually cumbersome parameter tuning processes and hence reduce their applicability to real-world deployments. In this paper, we discuss an efficient version of Kolmogorov-Arnold Network for audio-based speaker identification using three different types of audio inputs: bone conduction, air conduction, and throat microphone recordings, alongside evaluation on the publicly available LibriSpeech dataset to enhance reproducibility. To optimize the performance of these modalities, we investigate data-level, feature-level, and decision-level fusion techniques. An exhaustive comparison is performed against some of the prominent deep learning architectures, namely VGG-19, ResNet-152, EfficientNetV2, and DenseNet-121, both as standalone architectures and within multimodal combinations for the purpose of evaluating the proposed framework. We demonstrate that our model achieves a significant increase in accuracy up to 99.375% through the use of MFCC and GFCC feature inputs with data-level fusion through MFCC features and achieves that much accuracy using substantially fewer parameters than those of conventional deep learning models. This makes the KAN framework a highly efficient choice for audio-based speaker identification. One of the critical innovations within this design is a class of trainable activation functions that eliminate the need for traditional weight parameters, thereby providing scalable and computationally efficient solutions. This work contributes to the field of biometrics by providing a refined, resource-efficient approach to speaker identification based on multimodal audio data. © 2013 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Circuit Design using Quantum Dot technology 1

Circuit Design using Quantum Dot Technology

引用

1st International Conference on Intelligent Systems and Computational Networks, ICISCN 2025

作者： Manjunath, T.C. Kuravatti, Prakash Rajarajeswari College of Engineering Department of Computer Science & Engineering Bangalore India Rajeev Institute of Technology Department of Electronics and Communication Engineering Hassan India

ISBN: (纸本)9798331529246

In this paper, the harnessing QCA for High-Performance, Low-Power Arithmetic Circuits w.r.t. the focusing on Multipliers and Square Circuits is presented. The background in relation to this work is presented next. Quantum-dot Cellular Automata (QCA) is an emerging technology in nano-electronics that offers promising advantages over traditional CMOS (Complementary Metal-Oxide-Semiconductor) technology, including low power consumption, high operational speeds, and exceptional integration capabilities. These features position QCA as a potential successor to CMOS in various applications. Multipliers, critical components in digital signal processing, and square circuits, essential for mathematical computations, significantly impact system throughput and power efficiency. This research focuses on designing QCA-based multiplier and square circuits to leverage these advantages. Methods, Results, and Concluding Remarks w.r.t. the work is given here. The QCA-based multiplier was constructed using binary adders and AND gates, while a comparative analysis was conducted between QCA and CMOS technologies to evaluate performance. The evaluation considered parameters such as cell count, area, and energy dissipation, utilizing the QCA Designer tool for validation. Results demonstrate that QCA designs offer significant improvements in terms of reduced power dissipation and area requirements compared to CMOS. These findings highlight the potential of QCA technology to enhance the efficiency and integration of critical components in digital systems, marking it as a transformative approach in nano-electronics. © 2025 IEEE.

关键词： CMOS integrated circuits

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：