检索结果-内蒙古大学图书馆

International Symposium on Applied Machine Intelligence and Informatics (SAMI)

作者： László Lindenmaier Szilárd Aradi Tamás Bécsi Balázs Fekete Dept. of Control for Transportation and Vehicle Systems Budapest University of Technology and Economics Budapest Hungary Dept. of Control Engineering and Information Technology Budapest University of Technology and Economics Budapest Hungary

ISBN: (数字)9798350379365

ISBN: (纸本)9798350379372

Advanced driver assistance and automated driving systems rely on an enhanced perception, which requires a reliable fusion of heterogeneous data from multiple sensors. Track-to-track fusion is a commonly used architecture in automotive perception systems. However, track management is still under-discussed among other modules of track-to-track fusion, i.e., association and state fusion, despite its significant impact on the fidelity of the fused data. Track management is responsible for maintaining the fused track list by handling appearing and disappearing objects and, more importantly, determining if a sensor track was generated by a real target or false detections. In this paper, we compare different track management strategies using simulated and real data from a radar-camera sensor cluster, analyzing their false alarm filtering effectiveness.

关键词： Target tracking Sensor fusion Radar tracking Sensor systems Reliability Informatics Machine intelligence Intelligent sensors Automotive engineering Vehicles

来源：评论

学校读者我要写书评

暂无评论

Deepfake Detection of Singing Voices With Whisper Encodings

Deepfake Detection of Singing Voices With Whisper Encodings

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Sharma, Falguni Gupta, Priyanka Centre for Artificial Intelligence Banasthali Vidyapith Jaipur India Dept. of Communication and Computer Engineering The LNM Institute of Information Technology Jaipur India

ISBN: (纸本)9798350368741

The deepfake generation of singing vocals is a concerning issue for artists in the music industry. In this work, we propose a singing voice deepfake detection (SVDD) system, which uses noise-variant encodings of open-AI's Whisper model. As counter-intuitive as it may sound, even though the Whisper model is known to be noise-robust, the encodings are rich in non-speech information, and are noise-variant. This leads us to evaluate Whisper encodings as feature representations for the SVDD task. Therefore, in this work, the SVDD task is performed on vocals and mixtures, and the performance is evaluated in %EER over varying Whisper model sizes and two classifiers-CNN and ResNet34, under different testing conditions. © 2025 IEEE.

关键词： anti-spoofing Deepfake Detection Singing voice deepfake detection Transfer Learning Whisper

来源：评论

学校读者我要写书评

暂无评论

Geometrically Shaped Constellation in Short-Packet Visible Light Communications for IIoT Applications

引用

IEEE Internet of Things Journal 2025年第11期12卷 18220-18234页

作者： Guo, Jia-Ning Zhang, Jian Gong, Chen Li, Longguang Yang, Xu Zhou, Jing Chen, Ru-Han information engineering university Henan Province 450000 China The CAS Key Laboratory of Wireless-Optical Communications School of Information Science and Technology University of Science and Technology of China Hefei230027 China The SJTU Paris Elite Institute of Technology Shanghai Jiao Tong University Shanghai200240 China Sixty-Third Research Institute National University of Defense Technology Nanjing China Dept. Computer Science and Engineering Shaoxing University Shaoxing China

Due to the high data rates, license-free operations and inherent security, visible light communication (VLC) is highly valued in industrial Internet of Things (IIoT) applications. Short packet communication, which offers ultra-low latency and efficient resource utilization, is suitable for the low-latency and massive connectivity demand of the IIoT system. It adopts finite blocklength codewords for data transmissions. In this paper, we present a general framework of designing geometrically shaped constellations in short-packet VLC with peak and average intensity constraints for IIoT applications. By leveraging tools from large deviation theory, we first characterize the second-order asymptotics of the optimal constellation shaping region under aforementioned intensity constraints, which serves as a good performance metric for the best geometric shaping in finite blocklength. To further incorporate a sufficiently large coding gain and a nearly maximum shaping gain, we construct multidimensional constellations by the nested structure of Construction B lattices, where the constellation shaping is implemented by controlling the boundary of the embedded sublattice, i.e., a strategy called coarsely shaping and finely coding. Fast algorithms for constellation mapping and demodulation are presented as well. As an illustrative example, we present an energy-efficient 24-dimensional constellation design based on the Leech lattice, whose superiority over existing constellation designs is verified by numerical results. © 2014 IEEE.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Self Organized Anchor Group based Routing Protocol for IoT Initial Setup 17

Self Organized Anchor Group based Routing Protocol for IoT I...

引用

17th International Conference on COMmunication Systems and NETworkS, COMSNETS 2025

作者： Gupta, Naveen Kumar Aggarwal, Apeksha Yadav, Rama Shankar Kumar Nagaria, Rajendra Dr B R Ambedkar National Institute of Technology Jalandhar Center of Energy and Environment Dept. of Information Technology Jalandhar India Jaypee Institute of Information Technology Dept. of CSE and IT Noida India Motilal Nehru National Institute of Technology Allahabad Department of Computer Science and Engineering Prayagraj India Motilal Nehru National Institute of Technology Allahabad Department of Electronics and Comm. Engineering Prayagraj India

ISBN: (纸本)9798331531195

Geographical routing approach relies on local geographic information and performs multi-hop routing with the help of local decisions. The nodes are represented by 3D coordinates as some real IoT networks may be deployed in high buildings, hilly areas, underground or underwater. The geographical routing protocols follow the greedy path and suffer from void node problem (VNP). The traditional recovery policies follow a longer path around the boundary nodes of the void region which may go out of energy quickly. Thus, this paper proposes the concept of anchor group and presents an efficient anchor group routing protocol (EAGR) for IoT networks. The anchor group is treated as a sub-destination region where a group of nodes behaves like sub-destinations, which bypass the boundary nodes of the void to minimize the path length and divide the network traffic, in case of void in the network. The paper is also focused on considering the residual energy in the contention-based next-hop selection process. The experimental work has been carried out in the OMNET++ simulator's INET framework and observed that the proposed method is outperforming existing protocols. Moreover, it has reported a longer network lifetime and is able to survive longer than other existing geographical routing protocols on high load. © 2025 IEEE.

关键词： Routing protocols

来源：评论

学校读者我要写书评

暂无评论

Proposal for a Job Management Tool for Deep Learning in a Mixed Environment of Heterogeneous GPUs

Proposal for a Job Management Tool for Deep Learning in a Mi...

引用

Consumer technology (ICCT-Pacific), International Conference on

作者： Nobuyuki Ishii Aoi Kondo Hiroaki Koga Dept. Control and Information Systems Engineering National Institute of Technology Kurume College Fukuoka Japan

ISBN: (数字)9798331504120

ISBN: (纸本)9798331504137

In recent years, the use of GPUs for computation has become commonplace in deep learning. It is not uncommon for deep learning labs to procure GPU servers for computation without standardising specifications such as GPU architecture and VRAM capacity, resulting in a heterogeneous GPU server environment. Container technology is also widely used as a method of isolating software execution environments. Among them, Docker is the de facto standard for container platforms. In this paper, we have developed a management system that can allocate deep learning training jobs to GPU servers (workers) using container technology to absorb differences in execution environments in heterogeneous GPU server environments. In experiments, we confirmed that the proposed system can estimate the completion time of training for each worker and can detect Out-Of-Memory (OOM) of GPUs in advance. We were able to confirm in advance the “inference time” and the “possibility of OOM”, which are difficult to predict when performing deep learning. We have succeeded in obtaining the “predicted value of learning time” and “possibility of OOM” that engineers need to know in advance, and have shown that the proposed system is effective in deep learning tests.

关键词： Deep learning Training Graphics processing units Containers Software Servers Resource management Proposals Standards Testing

来源：评论

学校读者我要写书评

暂无评论

Single-Phase Open-Circuit Fault-Tolerant control of Three-Phase PMSM Drives 14

Single-Phase Open-Circuit Fault-Tolerant Control of Three-Ph...

引用

14th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2025

作者： Minato, Yuichiro Nakata, Yuki Itoh, Jun-Ichi Technology Development Center Research and Development Headquarters Murata Machinery Ltd Kyoto Japan Nagaoka University of Technology Dept. of Electrical Electronics and Information Engineering Niigata Japan

ISBN: (纸本)9798331516116

This paper proposes a Fault-tolerant control method for a standard three-phase permanent magnet synchronous motor (PMSM) under a single-phase open-circuit fault condition. The method for driving a three-phase PMSM from a standstill using a three-leg, two-level inverter configuration is proposed. Experimental results show that 65% of the motor's rated speed and 50% of its rated torque maintains. © 2025 IEEE.

关键词： Traction motors

来源：评论

学校读者我要写书评

暂无评论

Verification of the Effect of Label Modification on Neural Network Learning

Verification of the Effect of Label Modification on Neural N...

引用

Consumer technology (ICCT-Pacific), International Conference on

作者： Tsubasa Oka Hiroaki Koga Dept. Control and Information Systems Engineering National Institute of Technology Kurume College Fukuoka Japan

ISBN: (数字)9798331504120

ISBN: (纸本)9798331504137

This study focuses on the problem of subjective labeling that arises in the application of machine learning with neural networks to the medical field. The first problem is that of subjectivity, which arises because the labeling of medical data relies on expert knowledge and the process can be subjective. The second problem is the reduced information content of labels caused by discrete and quantitative labeling. If the intermediate values between the discrete values also have evaluative meaning, the detailed information is lost due to the discrete numerical labeling. We focus on such instability in labels and propose an approach to label modification using values inferred during neural network training. Specifically, we consider labels that are discrete, quantitative numbers as continuous numbers, compare the inferred value with the label value at each step in the learning phase, and compare the label value with the inferred value at each step in the learning phase. The update method is to give the label a sign-responsive variation. The goal is to modify each label using the value inferred by the inference made during learning in this way. As an experiment, we conducted an experiment on label modification under pseudo-conditions to verify its effectiveness. The results showed that when the amount of label modification was controlled by a constant, the generalization performance and the nature of label modification differed depending on the constant. In particular, generalization performance was improved when the constant of the label modification amount was determined so that the label value was separated from the inferred value at each step.

关键词： Training Neural networks Machine learning Labeling

来源：评论

学校读者我要写书评

暂无评论

SB-AGT: A stochastic beam search-enhanced attention-based Gumbel tree framework for drug-drug interaction extraction from biomedical literature

引用

Computers in Biology and Medicine 2025年 189卷 110011-110011页

作者： Saad, Eman Kishk, Sherif Ali-Eldin, Amr Saleh, Ahmed I. Communication and Information Technology Center Mansoura University Egypt Electronics and Communication Engineering Dept. Mansoura University Egypt Computer Engineering & Control Systems Dept. Mansoura University Egypt

Drug-Drug Interactions (DDI) and Chemical-Protein Interactions (CPI) detection are crucial for patient safety, as unidentified interactions may lead to severe Adverse Drug Reactions (ADRs). While extensive DDI and CPI information exists within biomedical literature, manual extraction by experts proves time-intensive and resource-demanding. This study presents SB-AGT: Stochastic Beam Search-enhanced Attention-based Gumbel Tree an innovative approach to automated DDI and CPI extraction through an attention-based architecture incorporating a modified Gumbel-Tree method with stochastic beam search optimization. SB-AGT leverages multi-head attention mechanisms and enhanced latent tree structures to effectively capture complex syntactic features and drug relationship patterns. Also, the methodology includes comprehensive preprocessing protocols and addresses dataset imbalance through hybrid sampling techniques. Experimental validation on the DDIExtraction 2013 and CHEMPROT datasets demonstrates the model's effectiveness, achieving precision, surpassing traditional approaches and showing competitive performance compared to contemporary pre-trained models. Through parametric analysis, we establish optimal beam search configurations that maximize the model's extraction accuracy. © 2025 Elsevier Ltd

关键词： Stochastic models

来源：评论

学校读者我要写书评

暂无评论

Dynamic Bidirectional Feature Enhancement Network for Thin Cloud Removal in Remote Sensing Images

引用

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 13281-13295页

作者： Wang, Yu Chen, Hao Zhang, Ye Li, Guozheng Harbin Institute of Technology Dept. of Information Engineering Harbin China

Existing thin cloud removal methods primarily rely on generative paradigms or discriminative paradigms. Generative paradigms often suffer from training instability, while discriminative paradigms exhibit insufficient feature representation, and their loss strategies lack physical consistency, resulting in suboptimal performance. To address these issues, we propose a dynamic bidirectional feature enhancement network for thin cloud removal in optical remote sensing images. Firstly, we design a multidimensional attention module comprising skip-recursive dilated convolution-based spatial attention module and frequency-domain-driven channel attention module. These modules efficiently capture long-range contextual dependencies and suppress cloud noise interference. Next, an adaptive local feature enhancement block is constructed using cross-fusion and adjacent feature propagation between dynamic convolutions, aimed at enhancing the ability of model to recover details. Then, we present a dynamic enhancement-based bidirectional information flow module to model the dynamic interaction between multi-task features, guiding detail recovery and feedback for optimized cloud removal features. Finally, we design a physics-aware joint loss function incorporating atmospheric light consistency constraints to ensure the physical authenticity of cloud-free images. Evaluation on the RICE and WHUS2-CR datasets demonstrates that the proposed method is superior to compared methods in thin cloud removal and can improve the performance of baseline. © 2008-2012 IEEE.

关键词： Adaptive optics

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Evaluation of Environmental Impact on the Performance of V2V Visible Light Communication 5

Comprehensive Evaluation of Environmental Impact on the Perf...

引用

5th IEEE Middle East and North Africa COMMunications Conference, MENACOMM 2025

作者： Al Hasnawi, Rasha Marghescu, Ion Al-Mudhafar, Amjed A University Politehnica of Bucharest Faculty of Electronics Telecommunications and Information Technology Dept. Telecommunications Bucharest Romania Al-Furat Al-AwsatTechnical University Engineering Technical College-Al Najaf Dept. Avionics Technologies Engineering Al Najaf Iraq

ISBN: (纸本)9798331519957

This paper investigates the performance of vehicle-To-vehicle visible light communication (V2V-VLC) systems under different outdoor conditions through MATLAB simulation. The study focuses on the signal-To-noise ratio (SNR) as it is affected by factors such as atmospheric variations, lateral displacement, ambient light interference, and receiver aperture size. The channel path loss is analyzed, and the simulation results are validated against experimental *** contributions of this work include a comprehensive evaluation of key environmental and design parameters, integration of theoretical models with real-world data, and recommendations for improving system performance. The results provide valuable insights into optimizing V2V-VLC systems and form the basis for future advances in intelligent transportation technologies. © 2025 IEEE.

关键词： MATLAB

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：