检索结果-内蒙古大学图书馆

Detecting Inaudible Voice Commands via Acoustic Attenuation by Multi-channel Microphones

IEEE Transactions on Dependable and Secure Computing 2024年 1-17页

作者： Ji, Xiaoyu Zhang, Guoming Li, Xinfeng Qu, Gang Cheng, Xiuzhen Xu, Wenyuan Electrical Engineering of Zhejiang University Hangzhou China School of Computer Science and Technology of Shandong University Qingdao China Department of Electrical and Computer Engineering at the University of Maryland College Park MD USA

DolphinAttacks (i.e., inaudible voice commands) modulate audible voices over ultrasounds to inject malicious commands silently into voice assistants and manipulate controlled systems (e.g., doors or smart speakers). Eliminating DolphinAttacks is challenging if ever possible since it requires to modify the microphone hardware. In this paper, we design EarArray, a lightweight method that can not only detect such attacks but also identify the direction of attackers without requiring any extra hardware or hardware modification. Essentially, inaudible voice commands are modulated on ultrasounds that inherently attenuate faster than the one of audible sounds. By inspecting the command sound signals via the built-in multiple microphones on smart devices, EarArray is able to estimate the attenuation rate and thus detect the attacks. We propose a model of the propagation of audible sounds and ultrasounds from the sound source to a voice assistant, e.g., a smart speaker, and illustrate the underlying principle and its feasibility. We implemented EarArray using two specially-designed microphone arrays and our experiments show that EarArray can detect inaudible voice commands with an accuracy of above 99% and recognize the direction of the attackers with an accuracy of 97.89% and can also detect the laser-based attack with an accuracy of 100%. IEEE

关键词： Microphones

来源：评论

学校读者我要写书评

暂无评论

Matching 5G Connected Vehicles to Sensed Vehicles for Safe Cooperative Autonomous Driving

引用

IEEE Network 2024年第3期38卷 227-235页

作者： Tang, Zuoyin He, Jianhua Yang, Kun Chen, Hsiao-Hwa Aston University School of Computer Science and Digital Technologies BirminghamB4 7ET United Kingdom Essex University School of Computer Science and Electronic Engineering ColchesterCO4 3SQ United Kingdom National Cheng Kung University Department of Engineering Science Tainan70101 Taiwan

5G connected autonomous vehicles (CAVs) help enhance perception of driving environment and cooperation among vehicles by sharing sensing and driving information, which is a promising technology to avoid accidents and improve road-use efficiency. A key issue for cooperation among CAVs is matching communicating vehicles to those captured in sensors such as cameras, LiDAR, etc.. Incorrect vehicle matching may cause serious accidents. While centimeter level positioning is now available for autonomous vehicles, matching connected vehicles to sensed vehicles (MCSV) is still challenging and has rarely been studied. In this paper, we are motivated to investigate the MCSV problem for 5G CAVs, propose and assess solutions for the problem to bridge the research gap. We formulate the MCSV problem and propose two MCSV approaches to support cooperative driving. The first approach is based on vehicle registration number (VRN), which is unique to identify a vehicle and can be shared among CAVs for MCSV. VRN is hashed before sharing to protect privacy, and will be compared to the shared one for vehicle matching. The second MCSV approach is based on visual features of vehicle's external views, which are shared with other CAVs and compared to those obtained from visual sensors to match the vehicles of interest. A new MCSV dataset is developed to assess the effectiveness of the proposed approaches. Experiment results show that both approaches are feasible and useful, and they achieve a very low false positive rate, which is critical for cooperative driving safety. © 1986-2012 IEEE.

关键词： Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

AI LipReader-Transcribing Speech from Lip Movements 6

AI LipReader-Transcribing Speech from Lip Movements

引用

6th IEEE International Conference on Emerging Smart Computing and Informatics, ESCI 2024

作者： Ajitha, D. Dutta, Disha Saha, Falguni Giri, Parus Kant, Rohan School of Computer Science and Engineering Vellore Institute of Technology Vellore Department of Software Systems Vellore India School of Computer Science and Engineering Vellore Institute of Technology Department of Database Systems Vellore India School of Computer Science and Engineering Vellore Institute of Technology Vellore Department of Information Security Vellore India

ISBN: (纸本)9798350306613

Lip Reading AI is a discipline that is rapidly changing and has numerous applications in security, accessibility and human-computer interaction. This paper proposes a model which combines Convolutional Neural Networks (CNNs) to capture spatial capabilities, Long Short-Term Memory (LSTM) networks to examine temporal dependencies, and an adaptive interest mechanism. Meticulous preprocessing of the MIRACL VC-l dataset addressing challenges including one of a kind lip moves and occlusions accompanied with the aid of transitioning this study effortlessly to LRS2 dataset to complement lexemic versatility is one of its key function. The effects verify its robustness throughout unique datasets with superior overall performance towards cutting-edge techniques. Ablation checks suggest the crucial significance of every element in phrases of improving lip analyzing accuracy. Our proposed model version additionally suggests flexibility in restricted and naturalistic language situations. © 2024 IEEE.

关键词： Attention Mechanism Convolutional Neural Networks (CNNs) Deep Learning Lip Reading Long Short-Term Memory (LSTM) LRS2 Dataset MIRACLVC-J Dataset

来源：评论

学校读者我要写书评

暂无评论

A Comparative Exploration of Time Series Models for Wild Fire Prediction 4

A Comparative Exploration of Time Series Models for Wild Fir...

引用

4th International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies, ICAECT 2024

作者： Sri Sowmya, V. Sasikala, D. Theetchenya, S. Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Engineering Chennai India Sona College of Technology Department of Computer Science and Engineering Salem India

ISBN: (纸本)9798350343670

Wildfires pose a great threat to human safety and property arising from both natural and human causes. According to a technical assessment by the Forest Survey of India more than 95% of fires are of anthropogenic origin. Wildfires erupt due to burning of fossils by local communities for crop rotation, camp fires without proper supervision etc. Climate change further elevates the risk by fostering dry conditions. Traditionally, wildfire prediction relied on statistical models and expert judgment. However, the emergence of Machine Learning (ML) and Deep Learning (DL) techniques has significantly improved the accuracy of forest fire prediction. The objective of this work is to prevent wildfires and save the ecosystem. In this work, LightGBM(Light Gradient Boosting Machine) and LSTM(Long Short-Term Memory) machine learning models are utilized to predict the forest fire. Both models exhibit high F1 scores of 97% and 95% accuracy in forest fire prediction, enabling the development of reliable prediction systems. The results of these ML-based models may aid in identifying highrisk areas, optimizing prevention measures, refining evacuation plans, and guiding firefighting efforts. © 2024 IEEE.

关键词： Deep Learning(DL) LightGBM LSTM Machine learning(ML) Wildfires

来源：评论

学校读者我要写书评

暂无评论

Virtual Keyboard:A Real-Time Hand Gesture Recognition-Based Character Input System Using LSTM and Mediapipe Holistic

引用

computer Systems science & engineering 2024年第2期48卷 555-570页

作者： Bijon Mallik Md Abdur Rahim Abu Saleh Musa Miah Keun Soo Yun Jungpil Shin Department of Computer Science and Engineering Pabna University of Science and TechnologyPabna6600Bangladesh School of Computer Science and Engineering The University of AizuAizuwakamatsuFukushima965-8580Japan School of Computer and Information Technology Ulsan CollegeDong-guUlsan44610South Korea

In the digital age,non-touch communication technologies are reshaping human-device interactions and raising security concerns.A major challenge in current technology is the misinterpretation of gestures by sensors and cameras,often caused by environmental *** issue has spurred the need for advanced data processing methods to achieve more accurate gesture recognition and *** study presents a novel virtual keyboard allowing character input via distinct hand gestures,focusing on two key aspects:hand gesture recognition and character input *** developed a novel model with LSTM and fully connected layers for enhanced sequential data processing and hand gesture *** also integrated CNN,max-pooling,and dropout layers for improved spatial feature *** model architecture processes both temporal and spatial aspects of hand gestures,using LSTM to extract complex patterns from frame sequences for a comprehensive understanding of input *** unique dataset,essential for training the model,includes 1,662 landmarks from dynamic hand gestures,33 postures,and 468 face landmarks,all captured in real-time using advanced pose *** model demonstrated high accuracy,achieving 98.52%in hand gesture recognition and over 97%in character input across different *** excellent performance in real-time testing underlines its practicality and effectiveness,marking a significant advancement in enhancing human-device interactions in the digital age.

关键词： Hand gesture recognition M.P.holistic open CV virtual keyboard LSTM human-computer interaction

来源：评论

学校读者我要写书评

暂无评论

Incorporation of Two-Fold Trust in Group Recommender System to Handle Popularity Bias

引用

SN computer science 2024年第2期5卷 259页

作者： Richa Bedi, Punam Gupta, Shivani Kumar, Ranjeet Department of Computer Science and Engineering Birla Institute of Technology Mesra Jharkhand Ranchi India Department of Computer Science University of Delhi Delhi New Delhi India School of Computer Science Engineering and Technology Vellore Institute of Technology Tamilnadu Chennai India School of Electronics Engineering Vellore Institute of Technology Tamilnadu Chennai India

In recent years, there has been a rise in work which focuses on various issues of recommender systems other than accuracy. Popularity bias is one factor that causes the list of recommendations to deviate from the user’s expectation and causes biased list to generate. This paper proposed an approach that handles the popularity bias in trust aware group recommender system using trust propagation. The proposed work considers two factors, novelty and serendipity, to evaluate the presence of bias. By incorporating trust, along with novelty, diversity, and unexpectedness, the diverse list of recommendation that maintains accuracy should be generated. Trust propagation combines trustworthy user’s opinion indirectly involves enhancing the accuracy. The trust propagation and randomness in trust propagation ensures the trade-off between the accuracy and bias of the recommendations. A prototype is developed using the tourism dataset and compared using traditional approach of recommendation using various accuracy measures and coverage. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Diversity Novelty Popularity bias Recommender system Relevancy

来源：评论

学校读者我要写书评

暂无评论

RBLA: Rank-Based-LoRA-Aggregation for Fine-Tuning Heterogeneous Models in FLaaS 31st

RBLA: Rank-Based-LoRA-Aggregation for Fine-Tuning Heterogen...

引用

31st International Conference on Web Services, ICWS 2024, Held as Part of the Services Conference Federation, SCF 2024

作者： Chen, Shuaijun Tavallaie, Omid Nazemi, Niousha Zomaya, Albert Y. School of Computer Science The University of Sydney Camperdown Australia Department of Engineering Science University of Oxford Oxford United Kingdom

ISBN: (纸本)9783031770715

Federated Learning (FL) is a promising privacy-aware distributed learning framework that can be deployed on various devices, such as mobile phones, desktops, and devices equipped with CPUs or GPUs. In the context of server-based Federated Learning as a Service (FLaaS), FL enables a central server to coordinate the training process across multiple devices without direct access to local data, thereby enhancing privacy and data security. Low-Rank Adaptation (LoRA) is a method that efficiently fine-tunes models by focusing on a low-dimensional subspace of the model’s parameters. This approach significantly reduces computational and memory costs compared to fine-tuning all parameters from scratch. When integrated with FL, particularly in a FLaaS environment, LoRA allows for flexible and efficient deployment across diverse hardware with varying computational capabilities by adjusting the local model’s rank. However, in LoRA-enabled FL, different clients may train models with varying ranks, which poses challenges for model aggregation on the server. Current methods for aggregating models of different ranks involve padding weights to a uniform shape, which can degrade the global model’s performance. To address this issue, we propose Rank-Based LoRA Aggregation (RBLA), a novel model aggregation method designed for heterogeneous LoRA structures. RBLA preserves key features across models with different ranks. This paper analyzes the issues with current padding methods used to reshape models for aggregation in a FLaaS environment. Then, we introduce RBLA, a rank-based aggregation method that maintains both low-rank and high-rank features. Finally, we demonstrate the effectiveness of RBLA through comparative experiments with state-of-the-art methods. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Tensorial Evolutionary Computation for Spatial Optimization Problems

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第1期5卷 154-166页

作者： Si-Chao, Lei Xiaolin, Xiao Yue-Jiao, Gong Yun, Li Jun, Zhang the School of Computer Science and Engineering South China University of Technology Guangzhou510006 China the Shenzhen Institute for Advanced Study University of Electronic Science and Technology of China Shenzhen610056 China the School of Computer Science South China Normal University Guangzhou510006 China the Department of Electrical and Electronic Engineering Hanyang University Seoul15588 Korea Republic of

—Spatial optimization problems (SOPs) refer to a class of problems where the decision variables require spatial organization. Existing methods based on evolutionary algorithms (EAs) fit conventional evolutionary operators by flattening the spatial representations of SOPs to one dimension, which hence loses crucial spatial structures of the potential solutions. To address this issue, this article proposes a tensorial evolutionary algorithm (TEA) by formulating a tensor space of the SOPs, where the population of candidate solutions is represented as a third-order tensor. Accordingly, we design a tensor-product crossover operator and a two-phase mutation consisting of a local convolutional mutation operator and a global Gaussian mutation operator in the tensor space. These operators are all implemented based on tensor operations. Under the consideration of spatial context on individual solutions, we devise three types of evolutionary tensors, namely, the crossover tensor, the convolutional mutant tensor, and the Gaussian mutant tensor for the three tensorial evolutionary operators correspondingly. The resultant TEA is tested using two typical real-world SOPs, namely, the facility layout problem and the image steganography problem. Comparative experimental studies validate the effectiveness and robustness of TEA. © 2024 IEEE.

关键词： Evolutionary algorithms

来源：评论

学校读者我要写书评

暂无评论

Survey on Video Security:Examining Threats,Challenges,and Future Trends

引用

computers, Materials & Continua 2024年第9期80卷 3591-3635页

作者： Ali Asghar Amna Shifa Mamoona Naveed Asghar Department of Artificial Intelligence Faculty of ComputingThe Islamia University of BahawalpurBahawalpur63100Pakistan School of Computer Science College of Science and EngineeringUniversity of GalwayGalwayH91 TK33Ireland

Videos represent the most prevailing form of digital media for communication,information dissemination,and ***,theirwidespread use has increased the risks of unauthorised access andmanipulation,posing significant *** response,various protection approaches have been developed to secure,authenticate,and ensure the integrity of digital *** study provides a comprehensive survey of the challenges associated with maintaining the confidentiality,integrity,and availability of video content,and examining how it can be *** then investigates current developments in the field of video security by exploring two critical research ***,it examine the techniques used by adversaries to compromise video data and evaluate their *** these attack methodologies is crucial for developing effective defense ***,it explores the various security approaches that can be employed to protect video data,enhancing its transparency,integrity,and *** compares the effectiveness of these approaches across different use cases,including surveillance,video on demand(VoD),and medical videos related to disease ***,it identifies potential research opportunities to enhance video data protection in response to the evolving threat *** this investigation,this study aims to contribute to the ongoing efforts in securing video data,providing insights that are vital for researchers,practitioners,and policymakers dedicated to enhancing the safety and reliability of video content in our digital world.

关键词： Attacks threats security services video manipulation video security

来源：评论

学校读者我要写书评

暂无评论

Countering Deepfakes using an Improved Advanced CNN and its Ensemble with Pretrained Models 3

Countering Deepfakes using an Improved Advanced CNN and its ...

引用

3rd International Conference on Electrical, Electronics, Information and Communication Technologies, ICEEICT 2024

作者： Mathur, Abeer Tejpal, Moulik Bhargava, Kshitiz Natarajan, Krishnaraj Singh, Manvendra Vellore Institute of Technology School of Computer Science & Engineering Department of Software Systems Vellore India Vellore Institute of Technology School of Computer Science & Engineering Department of Database Systems Vellore India

ISBN: (纸本)9798350369083

The extensive spread of DeepFake images on the internet has emerged as a significant challenge, with applications ranging from harmless entertainment to harmful acts like blackmail, misinformation, and spreading false propaganda. To tackle this issue, this paper introduces a sophisticated DeepFake detection model designed to identify and mitigate the increase of these deceptive images. The model architecture integrates an ensemble approach, combining the strengths of two pre-trained Convolutional Neural Network (CNN) models - MobileNet and Xception - with a novel CNN architecture, the Advanced CNN (ACNN). This rigorous validation process enabled the model to achieve a high accuracy rate of 97.89% in detecting DeepFakes. The successful implementation of this ensemble CNN approach demonstrates its effectiveness in distinguishing between real and fabricated imagery with high precision. This research makes a substantial contribution to the field of digital image forensics, offering a reliable tool for stakeholders across various sectors to identify and counteract the spread of DeepFake images online. © 2024 IEEE.

关键词： Generative adversarial 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：