检索结果-内蒙古大学图书馆

${\sf Img2Acoustic}$Img2Acoustic: A Cross-Modal Gesture Recognition Method Based on Few-Shot Learning

IEEE Transactions on Mobile Computing 2024年第3期24卷 1496-1512页

作者： Yongpan Zou Jianhao Weng Wenting Kuang Yang Jiao Victor C. M. Leung Kaishun Wu College of Computer Science and Software Engineering Shenzhen University Shenzhen China Faculty of Computer Science and Control Engineering Shenzhen University of Advanced Technology Shenzhen China Artificial Intelligence Research Institute Shenzhen MSU-BIT University Shenzhen China Information Hub Hong Kong University of Science and Technology Guangzhou China

Acoustic-based human gesture recognition (HGR) offers diverse applications due to the ubiquity of sensors and touch-free interaction. However, existing machine learning approaches require substantial training data, making the process time-consuming, costly, and labor-intensive. Recent studies have explored cross-modal methods to reduce the need for large training datasets in behavior recognition, but they typically rely on open-source datasets that closely align with the target domain, limiting flexibility and complicating data collection. In this paper, we propose ${\sf Img2Acoustic}$ , a novel cross-modal acoustic-based HGR approach that leverages models trained on open-source image datasets (i.e., EMNIST, Omniglot) to effectively recognize custom gestures detected via acoustic signals. Our model incorporates a task-aware attention layer (TAAL) and a task-aware local matching layer (TALML), enabling seamless transfer of knowledge from image datasets to acoustic gesture recognition. We implement ${\sf Img2Acoustic}$ on commercial devices and conduct comprehensive evaluations, demonstrating that our method not only delivers superior accuracy and robustness compared to existing approaches but also eliminates the need for extensive training data collection.

关键词： Training Gesture recognition Acoustics Data models Sensors Feature extraction Target recognition Mobile computing Image recognition Few shot learning

来源：评论

学校读者我要写书评

暂无评论

BioVLF-T: A Temporal Framework for Radiology Report Generation using Pre-Trained Vision Language Foundational Models

BioVLF-T: A Temporal Framework for Radiology Report Generati...

引用

Medical Imaging 2025: Clinical and Biomedical Imaging

作者： Kasturi, Akhil Vosoughi, Ali Hadjiyski, Nathan Wismüller, Axel Department of Electrical and Computer Engineering University of Rochester NY United States Department of Imaging Sciences University of Rochester NY United States Department of Computer Science University of Rochester NY United States Department of Biomedical Engineering University of Rochester NY United States Faculty of Medicine Institute of Clinical Radiology Ludwig Maximilian University Munich Germany

ISBN: (纸本)9781510685987

In this study, we propose BioVLF-T, a novel automatic radiology report generation framework built on a Bio-Vision Language Foundational Model (VLF) with a temporal framework. BioVLF-T enhances the contextual understanding of the vision-language model by leveraging historical patient data, incorporating previous chest X-rays and their corresponding radiology reports. We fine-tuned the BLIP-VQA, a vision-language foundational model, within this framework, integrating a CNN-based instructor model pre-trained on 14 pathological labels from CheXbert to accurately capture relevant pathologies. By sequentially integrating historical data, the model simulates a process akin to that of a radiologist, resulting in more precise and contextually relevant report generation. Our experimental results demonstrate significant performance enhancements, with the BioVLF-T framework showing a 120.47% improvement in BLEU scores and a 1069.50% increase in CIDEr scores over the conventional fine-tuning method. These improvements highlight the effectiveness of temporal data integration in radiology report generation. Additionally, we acknowledge the limitations of traditional metrics such as CIDEr in capturing the nuances of language and, in the future, will focus on scaling the model with larger datasets and exploring advanced evaluation methods, including LLM-based scores, to enhance prediction quality further. © 2025 SPIE.

关键词： Mammography

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional oscillation of an acoustic microbubble between two rigid curved plates

引用

Journal of Hydrodynamics 2021年第5期33卷 1019-1034页

作者： Kawa M.A.Manmi Imad A.Aziz Arun Arjunan Rostam K.Saeed Abdolrahman Dadvand Department ofMathematics College of ScienceSalahaddin University-ErbilKurdistan RegionIraq Department of Information Technology College of Engineering and Computer ScienceLebanese French UniversityKurdistan RegionIraq Faculty of Science and Engineering University of WolverhamptonWolverhamptonUK Faculty of Mechanical Engineering Urmia University of TechnologyUrmiaIran Karlsruhe Institute of Technology(KIT) Institute for Chemical Technology and Polymer ChemistryKarlsruheGermany

Understanding the near boundary acoustic oscillation of microbubbles is critical for the effective design of ultrasonic biomedical devices and surface cleaning ***,this study investigates the three-dimensional microbubble oscillation between two curved rigid plates experiencing a planar acoustic field using boundary integral method(BIM).The numerical model is validated via comparison with the nonlinear oscillation of the bubble governed by the modified Rayleigh-Plesset equation and with the axisymmetric model for an acoustic microbubble in infinite fluid ***,the influence of the wave direction and horizontal standoff distance(h)on the bubble dynamics(including jet velocity,jet direction,centroid movement,total energy,and Kelvin impulse)were *** was concluded that the jet velocity,the maximum radius and the total energy of the bubble are not significantly influenced by the wave direction,while the jet direction and the high-pressure region depend strongly on *** importantly,it was found that the jet velocity and the high-pressure region around the jet in acoustic bubble are drastically larger than their counterparts in the gas bubble.

关键词： Acoustic microbubble bubble dynamics boundary integral method(BIM) two curved plates potential flow

来源：评论

学校读者我要写书评

暂无评论

Prompt-Based Approach for Czech Sentiment Analysis

Prompt-Based Approach for Czech Sentiment Analysis

引用

2023 International Conference Recent Advances in Natural Language Processing: Large Language Models for Natural Language Processing, RANLP 2023

作者： Šmíd, Jakub Přibáň, Pavel Department of Computer Science and Engineering University of West Bohemia Faculty of Applied Sciences Czech Republic NTIS - New Technologies for the Information Society University of West Bohemia Faculty of Applied Sciences Czech Republic

ISBN: (纸本)9789544520922

This paper introduces the first prompt-based methods for aspect-based sentiment analysis and sentiment classification in Czech. We employ the sequence-to-sequence models to solve the aspect-based tasks simultaneously and demonstrate the superiority of our promptbased approach over traditional fine-tuning. In addition, we conduct zero-shot and few-shot learning experiments for sentiment classification and show that prompting yields significantly better results with limited training examples compared to traditional fine-tuning. We also demonstrate that pre-training on data from the target domain can lead to significant improvements in a zero-shot scenario. © 2023 Incoma Ltd. All rights reserved.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Detecting Parkinson's Disease using High-Dimensional Feature-Based Support Vector Regression

Detecting Parkinson's Disease using High-Dimensional Feature...

引用

2024 International Conference on Intelligent Algorithms for Computational Intelligence systems, IACIS 2024

作者： Aarthi, E. Jagadeesh, S. Lingineni, Sree Lakshmi Dharani, N.P. Pushparaj, T. Satyanarayana, A.N. Srm Institute of Science and Technology Faculty of Science and Humanities Department of Computer Science Kattankulathur India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Cse Chennai India R&d Institute of Science and Technology Vel Tech Rangarajan Dr Sagunthala Department of Mathematics Chennai India Department of Ece Tirupati India The American College Department of Information Technology Madurai India Sreenidhi Institute of Science and Technology Department of Ece Hyderabad India

ISBN: (纸本)9798350360660

Parkinson's disease (PD) is a progressive neurological disorder that significantly impacts patients' quality of life. Accurate and early detection of PD is crucial for effective management and treatment. A study proposes a High-Dimensional Feature-Based Support Vector Regression (HDFSVR) method for enhancing the accuracy of PD detection. This method integrates advanced normalization and feature selection techniques to improve diagnostic accuracy. Sensitivity analysis shows a prediction accuracy of 83%, while precision analysis indicates an accuracy of 87%. Recall analysis further highlights the method's efficacy with a prediction accuracy of 92.78%. The HDFSVR method achieves an accuracy of 94.78%, outperforming contemporary approaches such as Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), and Adaptive Genetic Weighted Optimizer (AGWO). It represents a substantial advancement in PD diagnostics, addressing the limitations of traditional support vector machines in handling high-dimensional data. The method can be seamlessly adopted in clinical settings, enhancing early and accurate detection of Parkinson's disease. © 2024 IEEE.

关键词： Support vector regression

来源：评论

学校读者我要写书评

暂无评论

Low-Resolution Face Recognition for CCTV and Edge-Powered Smart Attendance systems

Low-Resolution Face Recognition for CCTV and Edge-Powered Sm...

引用

IEEE Annual International computer Software and Applications Conference (COMPSAC)

作者： Nico Surantha Edward Yose Sani Muhamad Isa Department of Electrical Electronics and Communication Engineering Faculty of Science and Engineering Tokyo City University Tokyo Japan Computer Science Department BINUS Graduate Program-Master of Computer Science Bina Nusantara University Jakarta Indonesia

ISBN: (数字)9798350376968

ISBN: (纸本)9798350376975

Attendance systems have become more modern, and one of the biometric systems without physical contact is face recognition. However, many face-based attendance systems still carry out attendance individually and cannot detect multiple faces simultaneously. In addition, capturing facial data in real-time is still a challenge because the relatively large distance between the camera and the individual reduces the ability to recognize faces. The general solution is to use super-resolution to generate better-quality faces while maintaining the main facial recognition features. One technique still being researched is super-resolution generative adversarial networks (SRGAN). SRGAN can enlarge the resolution of captured images and maintain image quality sufficient for face recognition. The attendance system can be easily integrated into edge devices such as the Jetson Nano. This paper proposes automatic and effective attendance systems with the super-resolution technique to detect and recognize faces in low-resolution input. The experimental results show that using face data capture with a resolution of 40 × 40 pixels and a four-fold magnification results in a resolution of 160 × 160 pixels. Combining Face SRGAN with FaceNet architecture as the basis of face recognition can achieve an accuracy rate of 78.19% and an F1-Score of 81.13% with an average processing time of 1.61 seconds per frame on a PC and 14.55 seconds per frame on a Jetson Nano at an average of face recognition per frame of as many as up to 8 faces simultaneously.

关键词： Support vector machines Image quality Accuracy Face recognition Image edge detection Superresolution Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

Leveraging Artificial Intelligence in Online Educational Platforms to Enhance Continuous Professional Development in the Teaching Profession 2

Leveraging Artificial Intelligence in Online Educational Pla...

引用

2nd International Conference Computational and Characterization Techniques in engineering and sciences, IC3TES 2024

作者： Amankulovich, Khujakulov Sherzod Tokhirovich, Mannabov Javlon Bakhridinovich, Kadirov Tulkin Kuvandikovich, Taylakov Umid Ananth, Christo Kumar, T. Ananth Samarkand Regional National Center for Training Teachers in New Methods Uzbekistan Uzbekistan Samarkand Agroinnovations and Research University Uzbekistan Samarkand State University Uzbekistan&Adjunct Faculty Department of Electronics and Communication Engineering Vel Tech Rangarajan Dr.Sagunthala R&d Institute of Science and Technology Avadi Chennai India Ifet College of Engineering Computer Science and Engineering Tamilnadu India

ISBN: (纸本)9798350364699

To keep up with the dynamic nature of the modern classroom, teachers must have access to online learning platforms for professional development purposes. This investigation looks at how these platforms can be used to improve teachers' skills, make information sharing easier, and open doors to CPD opportunities. This study examines the relationship between online resources and methods to demonstrate how digital tools have an impact on teacher training and professional development by leveraging AI. It also draws attention to the fact that these platforms are helping with information-methodological support for teachers as well as the ongoing reforms in our nation. Incorporating empirical observations and case studies, the analysis demonstrates how effective online platforms are in fostering a collaborative and innovative environment for professional learning. Digital transformation has the potential to radically alter how teachers obtain professional development, and this study emphasizes that fact. This will enable teachers to effectively address the demands of modern classrooms. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Representation Convolutional Neural Network to Recognize Human Emotion 8

Multi-Representation Convolutional Neural Network to Recogni...

引用

8th International Conference on Information Technology and Digital Applications, ICITDA 2023

作者： Prabowo, Dwi Wahyu Setiawan, Noor Akhmad Debayle, Johan Nugroho, Hanung Adi Universitas Gadjah Mada Faculty of Engineering Department of Electrical and Information Engineering Yogyakarta Indonesia Universitas Darwan Ali Faculty of Computer Science Department of Information System Sampit Indonesia Centre Spin Mines Saint-Etienne Cnrs Umr 5307 Lgf Saint-Etienne France

ISBN: (纸本)9798350344691

Emotion recognition using brain-computer interfaces (BCIs) is an emerging field with the ongoing challenge of developing robust and efficient classification methods. This study introduces an original approach, the multi-representation convolutional neural network (MrCNN), designed to exploit the complementary features of recurrence plots (RP) and spectrograms. The efficacy of this new method was empirically evaluated and compared with the standalone RP and spectrogram methods in emotion classification tasks using electroencephalogram (EEG) signals. Two widely used datasets, DEAP and SEED, were used as the experimental platforms for this study. Employing a multifaceted analytical approach involving descriptive statistics, data visualization, and hypothesis tests, this study revealed that MrCNN significantly outperforms the standalone RP and spectrogram methods in terms of key performance metrics. Specifically, MrCNN achieved an accuracy of 0.9663, sensitivity of 0.9848, specificity of 0.9825, kappa score of 0.9511, and F1 score of 0.9656 for the DEAP dataset. For the SEED dataset, the accuracy was 0.9695, with sensitivity, specificity, kappa scores, and F1 score of 0.9844, 0.9842, 0.9543, and 0.9695, respectively. These results were statistically validated, and a comparative analysis with existing literature substantiated the robustness and superiority of the MrCNN method. This study sets new performance benchmarks and offers a reliable and effective approach for emotion classification using EEG signals. © 2023 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Real-Time Inertia Estimation and Validation Using PMU Data and the Swing Equation in Power systems

Real-Time Inertia Estimation and Validation Using PMU Data a...

引用

2025 IEEE Green Technologies Conference, GreenTech 2025

作者： Mhawesh, Mustafa A. Al-Sarray, Muthanna Mccann, Roy University of Arkansas Department of Electrical Engineering and Computer Science FayettevilleAR United States Al-Furat Al-Awsat Technical University Kufa Iraq Operation and Control Office Ministry of Electricity Baghdad Iraq

ISBN: (纸本)9798331520182

This paper presents a method to estimate power system inertia in real-time using Phasor Measurement Unit (PMU) data and the swing equation. Inertia is essential for maintaining power system stability during disturbances. The method calculates both electrical and mechanical power from PMU data and uses this information to estimate inertia by using Matlab. Then it validates the results by comparing calculated values with real measurements of the rate of change of frequency (ROCOF). Further validation is performed using data from a real-world generation trip event. The outcomes show that the proposed method is accurate, with minimal error, making it suitable for real-time applications. © 2025 IEEE.

关键词： Standby power systems

来源：评论

学校读者我要写书评

暂无评论

Performance LoRa Technology for Autonomous Vehicles 24

Performance LoRa Technology for Autonomous Vehicles

引用

24th International Seminar on Intelligent Technology and Its Applications, ISITIA 2023

作者： Adi, Puput Dani Prasetyo Purnama, Iwan Siregar, Arfanda Anugrah Juledi, Angga Putra Edi, Firman Karim, Abdul Wahyu, Yuyu Maulana, Fairuz Iqbal Susilo, Suhardi Atmoko Budi Toni Harahap, Ilham Maliki Patria, Lintang Bandung Indonesia Labuhanbatu University Faculty of Science and Technology North Sumatra Labuhanbatu Indonesia Politeknik Negeri Medan Telecommunication Engineering Study Program North Sumatra Medan City Indonesia Batam Institute of Technology Faculty of Industrial Technology Department of Engineering Management Riau Islands Batam City Indonesia Budi Darma University Department of Information Technology North Sumatra Medan Indonesia Bina Nusantara University School of Computer Science Computer Science Department Jakarta Indonesia University Labuhanbatu Faculty of Teacher Training and Education North Sumatra Labuhanbatu Indonesia Universitas Terbuka Faculty of Engineering Jakarta Indonesia

ISBN: (纸本)9798350313956

This research is an approach to intelligent vehicles with a LoRa communication system, LoRaWAN compatible for Long-Range and Outdoor Communication, but in this paper, we will test the ability of LoRa to handle autonomous vehicles, a Blind-Adaptive Data method will be tried on vehicles systems. The system offered is when the vehicle is running, the sensor will send position data in real-time, the number of vehicles is four, and GPS data is sent alternately using the Blind Adaptive Data Rate principle to avoid data collisions, and when the GPS shows a certain position, the value of SF is ascertained, which affects the data rate and ToA, and the output given is RSSI on each car, speed of a car, power consumption of sensor nodes, signal power, SNR, Bit rate or data rate, and sensitivity. With the Blind ADR principle, it is ensured that there is no data collision between car position data that causes packet loss. © 2023 IEEE.

关键词： Autonomous vehicles

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：