检索结果-内蒙古大学图书馆

32nd ACM International Conference on Multimedia, MM 2024

作者： Hai, Xuan Liu, Xin Tan, Yuan Liu, Gang Li, Song Niu, Weina Zhou, Rui Zhou, Xiaokang School of Information Science and Engineering Lanzhou University Lanzhou China The State Key Laboratory of Blockchain and Data Security Zhejiang University Hangzhou China School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Faculty of Business Data Science Kansai University Osaka Japan

ISBN: (纸本)9798400706868

Voice is one of the most widely used media for information transmission in human society. While high-quality synthetic voices are extensively utilized in various applications, they pose significant risks to content security and trust building. Numerous studies have concentrated on AI-synthesized voice detection to mitigate these risks, with many claiming to achieve promising performance. However, recent research has demonstrated that fake voice detectors suffer from serious overfitting to speaker-irrelative features (SiFs) and cannot be used in real-world scenarios. In this paper, we analyze the limitations of existing fake voice detectors and propose a new design philosophy, guiding the detection model to prioritize learning human voice features rather than the difference between the human voice and the synthetic voice. Based on this philosophy, we propose a novel AI-synthesized voice detection framework named SiFSafer, which uses pre-trained speech representation models to enhance the learning of feature distribution in human voices and the adapter fine-tuning to optimize the performance. The evaluation shows that the average EERs of existing fake voice detectors in the ASVspoof datasets can exceed 20% if the SiFs like silence segments are removed, while SiFSafer achieves an EER of less than 8%, indicating that SiFSafer is robust to SiFs and strongly resistant to existing attacks. © 2024 ACM.

关键词： Fake detection

来源：评论

学校读者我要写书评

暂无评论

Video integrity verification based on blockchain 2

Video integrity verification based on blockchain

引用

2nd IEEE International Conference on electronic engineering, ICEEM 2021

作者： Kamal, Randa Hemdan, Ezz El-Din El-Fishway, Nawal Menoufia University Department of Computer Science and Engineering Faculty of Electronic Engineering Menouf Egypt

ISBN: (纸本)9781665418423

with the existence of the new revolution of software techniques, tampering with the videos became easier than ever. For retaining the integrity of the captured scenes, novel methods are required. Blockchain is now one of the most interesting integrity verification techniques. Blockchain provides trust among the participant nodes;"users"without the need of a trusted third party. In this paper, we propose a new blockchain-based framework for integrity verification of videos. In the proposed framework, the scene is captured and hashed in its original state and the captured frame;"video frame"is stored as a reference frame. The hash value, frame name, and timestamp of the capturing time are stored in a CSV file. The genesis block of a permissioned blockchain is created with this data. Other frames are captured for the same scene every predefined period and hashed. If the new hash value is identical to the reference hash, then the new frame is deleted to preserve storage space, and the metadata of the frame is stored in the same CSV file. If the hash values are not identical, then the name of the frame is changed to the timestamp of the capturing time, the frame then is propagated to all participant nodes, an email is sent to a predefined mail to take the appropriate action, and finally, a new block will be created with the new metadata. The proposed system ensures the integrity of the captured videos, ease to use and robustness. © 2021 IEEE.

关键词： Blockchain

来源：评论

学校读者我要写书评

暂无评论

The effect of graph complexity in an energy-based FDI approach

引用

IFAC-PapersOnLine 2024年第4期58卷 622-627页

作者： Jan H. Smith George van Schoor Kenneth R. Uren Thomas Schäfer School of Electrical Electronic and Computer Engineering Faculty of Engineering North-West University South Africa Unit for Energy and Technology Systems Faculty of Engineering North-West University Institute for Process Technology Process Automation and Measurement Technology Zittau/Görlitz Germanuy

This paper introduces a structured approach for creating node signature graphs in energy graph-based visualization for industrial systems, focusing on anomaly detection. It examines the relationship between graph cardinality and the singular value decomposition analysis applied in the energy graph-based visualization method. The results reveal a trade-of between graph complexity and singular value resolution. The study highlights how graph structure affects energy flow monitoring and anomaly detection. It notably enables fault detection in components not represented as nodes, enhancing the methodology in energy-based system monitoring.

关键词： Fault detection isolation Graph reduction Graph matching Hybrid FDI Industrial process Thermal flow process Graph-based anomaly detection

来源：评论

学校读者我要写书评

暂无评论

New In-Air Signature Datasets

New In-Air Signature Datasets

引用

2023 International Symposium on Networks, computers and Communications, ISNCC 2023

作者： Alobaidi, Abdulazeez R. Dhieb, Thameur Nuimi, Zeina N. Hamdani, Tarek M. Wali, Ali Alimi, Adel M. University of Sfax REGIM-Lab.: REsearch Groups in Intelligent Machines Enis Tunisia Imam ja'Afar Al-Sadiq University Faculty of Information Technology Department of Computer Techniques Engineering Iraq Higher Institute of Business Administration University of Gafsa Tunisia College of Engineering Department of Electrical Engineering University of Misan Iraq Higher Institute of Computer Science Mahdia University of Monastir Tunisia Higher Institute of Computer Science and Multimedia of Sfax University of Sfax Tunisia University of Johannesburg Faculty of Engineering and the Built Environment Department of Electrical and Electronic Engineering Science South Africa

ISBN: (纸本)9798350335590

Compared to traditional biometric systems, in-air signatures are considered more robust and secure than classical pen paper. A few datasets capturing in-air signatures have been introduced, utilizing various devices such as the Leap Motion and the Microsoft Kinect sensor camera. However, these devices are not exempt from shortcomings and exhibit certain limitations. The expenses associated with their implementation and the requirement for technical proficiency in operating them present notable challenges for in-air signature analysis. Additionally, users may encounter difficulties in adapting their finger movements to fit within the device's limited field of view, particularly if they lack familiarity with these devices. To address these concerns, this paper proposes the creation of three in-air signature datasets using solely the camera of a laptop or a smartphone, eliminating the need for any additional specialized equipment. Our datasets were collected in three ways. The first is the In-Air Signature dataset (IAS dataset) and the second is the In-Air Signature dataset using a transparent Glass Plate (IASGP dataset) while the third is the In-Air Signature dataset using Smart Phone (IASSP dataset). Forty volunteers participated in the construction of these datasets. Their ages ranged from 21 to 40 years. Each volunteer signs in the air five signatures and imitates five signatures of five other volunteers. Our in-air signatures datasets are publicly available and can be used for various research tasks like in-air signature verification and identification. © 2023 IEEE.

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

A∗ Algorithm-Based Indoor Navigation System for Multi-Story Building 13

A∗ Algorithm-Based Indoor Navigation System for Multi-Story...

引用

13th IEEE Global Conference on Consumer electronic, GCCE 2024

作者： Sariman, Muhammad Shazmin Othman, Maisara Saon, Sharifah Mahamad, Abd Kadir Kadir, Herdawatie Abdul Ahmadon, Mohd Anuaruddin Yamaguchi, Shingo Universiti Tun Hussein Onn Malaysia Faculty of Electrical and Electronic Engineering Johor Malaysia Universiti Teknologi Petronas Department of Computer and Information Sciences Perak Malaysia Yamaguchi University Graduate School of Sciences and Technology for Innovation Japan

ISBN: (纸本)9798350355079

Augmented reality (AR)-based indoor guidance is gaining popularity in various industries. With AR's capability to connect to other devices and perform innovative tasks, technological firms are investing heavily in its development. At the beginning of each semester, university students often struggle to navigate the campus, relying on maps or seeking help from others. This project addresses the need for an AR-based indoor navigation app specifically for Block QB, faculty of Electrical and electronic engineering (FKEE), Universiti Tun Hussein Onn Malaysia (UTHM). The proposed method addresses a shortest path planning problem in a multi-story building in a computationally efficient way by adopting an A∗ algorithm. The app incorporates the A∗ algorithm to ensure users can find the fastest and most accurate routes. The indoor navigation app was successfully developed for Android users utilizing the 3D Unity software. It gives users with real-time, precise guidance within the three-story Block QB FKEE building. © 2024 IEEE.

关键词： Motion planning

来源：评论

学校读者我要写书评

暂无评论

DDoS Detection Using Active and Idle Features of Revised CICFlowMeter and Statistical Approaches 4

DDoS Detection Using Active and Idle Features of Revised CIC...

引用

4th International Conference on Advanced Science and engineering, ICOASE 2022

作者： Ali, Basheer Husham Sulaiman, Nasri Al-Haddad, S.A.R. Atan, Rodziah Hassan, Siti Lailatul Mohd Department of Electrical and Electronic Engineering Serdang Malaysia Faculty of Engineering Department of Electrical and Electronic Engineering Serdang Malaysia Department of Computer and Communication Systems Engineering Serdang Malaysia Software Eng. and Information Systems Dept Serdang Malaysia School of Electrical Engineering Universiti Teknologi Mara Shah Alam Malaysia

ISBN: (纸本)9781665472227

Distributed Denial of services (DDoS) attack is one of the most dangerous attacks that targeted servers. The main consequence of this attack is to prevent users from getting their legitimate services by bringing down targeted victim. CICFlowMeter tool generates bi-directional flows from packets. Each flow generates 83 of different features. The research focuses on 8 features which are active min (f1), active mean (f2), active max (f3), active std (f4), idle min (f5), idle mean (f6), idle max (f7), and idle std (f8). CICFlowMeter tool has several problems that affected on the detection accuracy of DDoS attacks. The idle and active based feature of Shannon entropy and sequential probability ratio test (SE-SPRT) approach was implemented in this research. The problems of original CICFlowMeter were presented, and the differences between original and revised version of CICFlowMeter tool were explored. The DARPA database and confusion matrix were used to evaluate the detection technique and present the comparison between two versions of CICFlowMeter. The detection method detected neptune and smurf attacks and had higher accuracy, f1-score, sensitivity, specificity, and precision when revised version of CICFlowMeter used to generate flows. However, the detection method failed to detect neptune attack and had higher miss-rate, lower accuracy, lower f1-score, and lower specificity, and lower precision when original version used in generating flows. © 2022 IEEE.

关键词： Denial-of-service attack

来源：评论

学校读者我要写书评

暂无评论

Wildlife Intrusion Detection System for Agriculture Protection Based on Image Recognition 13

Wildlife Intrusion Detection System for Agriculture Protecti...

引用

13th IEEE Global Conference on Consumer electronic, GCCE 2024

作者： Kiat, Yong Kit Som Mahamad, Abd Kadir Saon, Sharifah Othman, Maisara Kadir, Herdawatie Binti Abdul Ahmadon, Mohd Anuaruddin Yamaguchi, Shingo Universiti Tun Hussein Onn Malaysia Faculty of Electrical and Electronic Engineering Johor Malaysia Universiti Teknologi Petronas Department of Computer and Information Sciences Perak Malaysia Yamaguchi University Graduate School of Sciences and Technology for Innovation Japan

ISBN: (纸本)9798350355079

The proposed project uses a Raspberry Pi microcontroller to prevent crop losses caused by animals like dog, wild pigs, and monkeys. These animals pose a significant threat to farmers, leading to financial losses. This project offers an innovative solution, as it is impractical to physically secure entire fields or have farmers present at all times. Raspberry Pi 4 and the Raspberry Pi Camera Module 2 NoIR havebeen used in this project, to captures images of intruding animals, sends notifications to users, and triggers an alarm. The model obtained an accuracy of 95% for intruder detection. The goal is to provide timely alerts, allowing farmers to respond quickly and minimize crop damage, effectively addressing wildlife-induced agricultural losses. © 2024 IEEE.

关键词： Livestock

来源：评论

学校读者我要写书评

暂无评论

A new dataset and systematic evaluation of deep learning models for student activity recognition from classroom videos

A new dataset and systematic evaluation of deep learning mod...

引用

2022 International Conference on Multimedia Analysis and Pattern Recognition, MAPR 2022

作者： Nguyen, Phuong-Dung Nguyen, Hong-Quan Nguyen, Thuy-Binh Le, Thi-Lan Tran, Thanh-Hai Vu, Hai Huu, Quynh Nguyen Hanoi University of Science and Technology School of Electrical and Electronic Engineering Hanoi Viet Nam Viet-Hung Industrial University Faculty of Information Technology Hanoi Viet Nam University of Transport and Communications Faculty of Electrical-Electronic Engineering Hanoi Viet Nam Thuyloi University Faculty of Computer Science and Engineering Hanoi Viet Nam

ISBN: (数字)9781665474108

ISBN: (纸本)9781665474108

Student activity, a non-verbal modality, is an important cue to imply the students' learning state and engagement level during class. Quantitative observation of the student activities will contribute to innovating teaching methods. However, traditional methods that rely on manual observation are time-consuming. Nowadays, surveillance cameras have been widely deployed in the classroom. This results in a massive amount of classroom videos. However, the exploitation of these videos is limited due to the lack of annotated information. This paper has two main contributions. First, a fully annotated dataset dedicated to student activity analysis has been proposed. To this end, five activities in the classroom context including sitting, standing, sleeping, raising-hand and using-phone have been defined and collected in different class sessions. Second, three deep learning models have been fully assessed for the activities recognition. © 2022 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

A High-Performance MTJ-LUT Circuit Using 4T1M Architecture 17

A High-Performance MTJ-LUT Circuit Using 4T1M Architecture

引用

17th IEEE International Conference on Solid-State and Integrated Circuit Technology, ICSICT 2024

作者： Pan, Yu Zhang, Yuejun Guo, Shuaicheng Tian, Yuanxin Hong, Bo Fang, Rui Wen, Liang Ding, Yong Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China China Coast Guard Academy Department of Electronic Technology Ningbo315211 China College of Integrated Circuits Zhejiang University Hangzhou310058 China

ISBN: (纸本)9798350361834

As fundamental components of programmable logic circuits, Lookup Table (LUT) circuits enable the implementation of arbitrary combinational logic. The volatility, standby power dissipation, and propagation delay of LUT circuits critically affect the performance of programmable logic circuits. In order to address the problems of circuit volatility, propagation delay and standby power dissipation, a high-performance Magnetic Tunnel Junction LUT (MTJ-LUT) circuit design scheme is proposed based on the 4T1M structure. According to the characteristics of MTJ principle, and comparing with the classical MTJ-LUT memory cell, MTJ-LUT memory module based on 4T1M structure is proposed;Based on the LUT circuit principle of 4T1M structure, a circuit architecture with current mirror bias circuitry is proposed to adopt the compliant reference resistor, and the steady-current-powered reading mode is used to achieve the optimal reading operation and reduce the delay;The circuit function is implemented in SMIC 40nm process using SPICE simulation tool. The experimental results show that the propagation delay and power dissipation of the MTJ-LUT circuit proposed are reduced by 72% and 15%, respectively, compared with the conventional MTJ-LUT circuit. © 2024 IEEE.

关键词： Delay circuits

来源：评论

学校读者我要写书评

暂无评论

Identification of Haploid Maize Seeds Using Efficient Channel Attention with Selective Kernel Networks and Multi-layer Stacked Ensemble

Identification of Haploid Maize Seeds Using Efficient Channe...

引用

Cognitive Computing and Complex Data (ICCD), International Conference on the

作者： Lieto Wamunyima Haiyi Bian Bobwillie Kessellie Sulonteh Faculty of Electronic Information Engineering Huaiyin Institute of Technology Huai’an City China Faculty of Computer Application and Engineering RIMT University Mandi Gobindgarh Punjab India

ISBN: (数字)9798350352894

ISBN: (纸本)9798350352900

This study focuses on the recognition of maize haploid seeds using convolutional neural networks (CNNs). A dataset comprising 3,000 maize seed images, consisting of 1,230 haploid maize seed images and 1,770 diploid maize seed images, was utilized for experimentation. Classical CNN models, including VGG, AlexNet, ResNet, and SKNet, were employed to compare their effectiveness in haploid maize seed recognition. Additionally, this study proposes the ECA-SKEnsemble model, which enhances the SKNet architecture by replacing its fully connected layer with a one-dimensional convolution, thus reducing model complexity. The ECA-SKEnsemble integrates multiple instances of the ECA-SKNet model, each trained with different initializations and training techniques. By combining these models using ensemble techniques such as weighted voting or averaging, ECA-SKEnsemble achieves superior performance in maize seed recognition. The model attained an accuracy of 93.04%, outperforming standard CNN architectures, including VGG and ResNet. This work demonstrates the effectiveness of ensemble learning combined with efficient channel attention for improving maize seed recognition accuracy.

关键词： Training Accuracy Convolution Computational modeling computer architecture Convolutional neural networks Ensemble learning Kernel Tuning Standards

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：