检索结果-内蒙古大学图书馆

6D Pose Estimation and Correction for Fully Occluded 3D Model X-Ray Source Using Deeplabv3 Utilizing Mobilenetv2 Backbone and Dense Fusion Algorithm

6D Pose Estimation and Correction for Fully Occluded 3D Mode...

引用

International Symposium on Networks, computers and Communications (ISNCC)

作者： Jayson P. Rogelio Elmer Jose P. Dadios Argel Bandala Raouf Naguib Ryan Rhay Vicerra Edwin Sybingco laurence Gan Lim Department of Science and Technology Metals Industry Research and Development Center De La Salle University Philippines De La Salle University College of Manufacturing and Management Manila Philippines De La Salle University College of Electronics and Communications Engineering Manila Philippines Mathematics Computer Science and Engineering Liverpool Hope University Liverpool United Kingdom De La Salle University College of Mechanical Engineering Manila Philippines

ISBN: (数字)9798350364910

ISBN: (纸本)9798350364927

Recent advancements in computer vision, notably in 6D pose estimation with methods like Deeplabv3 and Dense Fusion, show promising results. However, a substantial research gap exists in addressing the limitations of these methods in occluded scenarios, which in need further research to enhance their effectiveness and adaptability in situations involving hidden or partially obscured objects. This study introduces a novel approach employing 6D object pose estimation and pose correction by integrating DeeplabV3 and modified dense fusion. The study exhibits enhanced accuracy in pose estimation and correction for occluded 3D object model x-ray source. Evaluation metrics, including intersection-over-union and mean average, demonstrate high accuracy percentages for detecting the body (98.91%), handle (96.57%), and aperture (89.54%). The mean IoU for each part of the 3D model portable X-ray source ranges from 65.69% to 76.42%. Pose estimation accuracy, assessed through the Average Distance Difference (ADD) metric, indicates superior performance for static pose estimation closer to the camera. Dynamic pose estimation exhibits higher average ADD metrics in scenes with total occlusion. The robustness metric reveals lower lost tracking counts in scenes without occlusion, emphasizing the algorithm's challenges in fully occluded scenarios.

关键词： Measurement Solid modeling Accuracy Three-dimensional displays Heuristic algorithms Pose estimation Cameras Visual servoing Robustness Iterative algorithms

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach for Work Related Stress Detection from Audio Streams in Cyber Physical Environments 25

A Deep Learning Approach for Work Related Stress Detection f...

引用

25th IEEE International Conference on Emerging Technologies and Factory Automation, ETFA 2020

作者： Madhavi, Ishara Chamishka, Sadil Nawaratne, Rashmika Nanayakkara, Vishaka Alahakoon, Damminda De Silva, Daswin University of Moratuwa Department of Computer Science Engineering Sri Lanka La Trobe University Centre for Data Analytics and Cognition Australia

ISBN: (纸本)9781728189567

Work-related stress is an uncompromising burden with compounding effects on individuals, communities, organizations, and the economy. Highly automated digital environments, such as smart factories and cyber-physical ecosystems, are significantly impacted by these negative effects due to the inherent constraints on human engagement and social interaction. Verbal communication between human operators is a critical resource in such environments, as it can be effectively utilized to address this challenge. A mix of statistical and artificial intelligence (AI) techniques have been reported in recent literature for the detection of stress-related indicators from audio recordings. However, none of these studies has focused on the challenges of detecting work-related stress in cyber-physical environments, such as smart factories, where verbal communication is not only constrained but also impaired by background noise and other disturbances common to factory settings. In this paper, we address these challenges by proposing a novel deep learning approach, based on the workings of the Convolutional Neural Network (CNN) and Growing Self-Organizing Map (GSOM) algorithms. In addition to this novel AI approach, sampling strategy and speaker diarization techniques are utilised for noise reduction. Pitch and speech augmentation techniques address the data imbalance issue common in most real-world datasets. The accuracy and effectiveness of the proposed approach is demonstrated using a benchmark dataset (DAIC-WOZ). We report F1 scores of 82% and 64% for normal and distressed classes respectively, which outperform the state-of-the-art models. We conclude with a discussion on the empirical evaluation of the proposed approach in cyber-physical environments and directions for future work. © 2020 IEEE.

关键词： Cyber Physical System

来源：评论

学校读者我要写书评

暂无评论

A 3-Phase Supercapacitor Based Single-Source MLI with Self-Voltage Balancing and Boosting Ability

A 3-Phase Supercapacitor Based Single-Source MLI with Self-V...

引用

International Conference on Applied Superconductivity and Electromagnetic Devices, ASEMD

作者： Sudipto Mondal Shuvra Prokash Biswas Md. Sabbir Bin Islam Md. Rabiul Islam Afef Fekih Department of Electronics & Telecommunication Engineering Rajshahi University of Engineering & Technology Rajshahi Bangladesh Bangladesh Army University of Science and Technology Saidpur Bangladesh School of Electrical Computer & Telecommunications Engineering University of Wollongong NSW Australia Dept. of Electrical and Computer Engineering University of Louisiana at Lafayette Lafayette LA USA

This paper presents a new 3-phase multilevel inverter (MLI) based on a supercapacitor (SC) unit, which is ideal for medium and low-voltage grid-feeding photovoltaic (PV) applications since it is built to deliver nine different output voltage (line-to-line) levels from a single isolated dc source. Two SC units are utilized in each phase to create virtual dc sources while providing self-voltage balancing for the SCs. As a result, no extra control complexity is required compared to conventional neutral point clamped (NPC), flying capacitor (FC) or any other series connected capacitor-based inverter topologies. The proposed MLI is designed such a way that, the SCs could boost the output voltage up to four times than the input dc voltage. The MATlaB/Simulink computer simulation environment and a preliminary experimental prototype setup are used to verify the performance of the proposed MLI.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Design and Development of a Real-time Monitoring System for ACL Injury Prevention

Design and Development of a Real-time Monitoring System for ...

引用

International Conference on Networking, Embedded and Wireless Systems (ICNEWS)

作者： R Harinandan Mahesh Kumar N Pratyush Puram Vamshi Chennagiri Rajarao Padma Krutthika Hirebasur Krishnappa Jahnavi Rohini Raghunandan Department of Electronics and Communication Engineering Dayananda Sagar College of Engineering Bengaluru India Department of Medical Electronics Engineering Dayananda Sagar College of Engineering Bengaluru India Computer Science Department Southern University and A&M College Baton Rouge Louisiana LA United States

ISBN: (数字)9798350367461

ISBN: (纸本)9798350367478

Anterior Cruciate Ligament (ACL) injuries are common in sports and can have long-term effects on the injured party. They frequently require intensive rehabilitation. The creation, assessment, and proof of concept of an electronic monitoring tool designed to identify and stop ACL injuries are presented in this work. In later stages, we use machine learning algorithms to analyse biomechanical data in real-time, giving users immediate feedback on movement patterns such as slowing down walking pace or prohibiting the user from moving at a certain angle to avoid any sort of injury risk factors. The device makes use of wearable sensors, such as an MPU6050 (accelerometer/gyroscope) sensor and an Electromyography (EMG) sensor. Through a series of laboratory tests and field trials, the effectiveness and usability of the device were assessed, demonstrating its potential to improve athlete performance, prevent injuries, and enhance rehabilitation outcomes. According to the findings, there is hope for reducing the risk of ACL injuries and improving sports training regimens with the use of an electronic monitoring device.

关键词： Machine learning algorithms Prevention and mitigation Real-time systems Electromyography Ligaments Biomedical monitoring Injuries Wearable sensors Monitoring Sports

来源：评论

学校读者我要写书评

暂无评论

Emerging trends in biomedical trait-based human identification: A bibliometric analysis

引用

SlaS Technology 2024年第3期29卷 100136-100136页

作者： Girdhar, Nancy Sharma, Deepak Kumar, Rajeev Sahu, Monalisa Lin, Chia-Chen L3i University of La Rochelle La Rochelle 17000 France Department of Computer Science Christian-Albrechts-University zu Kiel Kiel 24118 Germany Blockchain Technology Research Lab Department of Computer Science and Engineering Delhi Technological University New Delhi 110042 India Amrita School of Computing Amrita Vishwa Vidyapeetham 522503 Andhra Pradesh India Department of Computer Science and Information Engineering National Chin-Yi University of Technology Taichung 411030 Taiwan

Personal human identification is a crucial aspect of modern society with applications spanning from law enforcement to healthcare and digital security. This bibliometric paper presents a comprehensive analysis of recent advances in personal human identification methodologies focusing on biomedical traits. The paper examines a diverse range of research articles, reviews, and patents published over the last decade to provide insights into the evolving landscape of biometric identification techniques. The study categorizes the identified literature into distinct biomedical trait categories, including but not limited to, fingerprint and palmprint recognition, iris and retinal scanning, facial recognition, voice and speech analysis, gait recognition, and DNA-based identification. Through systematic analysis, the paper highlights key trends, emerging technologies, and interdisciplinary collaborations in each category, revealing the interdisciplinary nature of research in this field. Furthermore, the bibliometric analysis examines the geographical distribution of research efforts, identifying prominent countries and institutions contributing to advancements in personal human identification. Collaboration networks among researchers and institutions are visualized to depict the knowledge flow and collaborative dynamics within the field. Overall, this study serves as a valuable reference for researchers, practitioners, and policymakers, shedding light on the current status and potential future directions of personal human identification leveraging biomedical traits. Copyright © 2024 The Author(s). Published by Elsevier Inc. All rights reserved.

关键词： Bibliometric Biometric identification Deep learning Machine learning Pattern recognition Traits analysis

来源：评论

学校读者我要写书评

暂无评论

Discovering Mixtures of Structural Causal Models from Time Series Data

arXiv

引用

arXiv 2023年

作者： Varambally, Sumanth Ma, Yi-An Yu, Rose Halıcıoğlu Data Science Institute University of California La Jolla San Diego United States Department of Computer Science and Engineering University of California La Jolla San Diego United States

Discovering causal relationships from time series data is significant in fields such as finance, climate science, and neuroscience. However, contemporary techniques rely on the simplifying assumption that data originates from the same causal model, while in practice, data is heterogeneous and can stem from different causal models. In this work, we relax this assumption and perform causal discovery from time series data originating from a mixture of causal models. We propose a general variational inference-based framework called MCD to infer the underlying causal models as well as the mixing probability of each sample. Our approach employs an end-to-end training process that maximizes an evidence-lower bound for the data likelihood. We present two variants: MCD-Linear for linear relationships and independent noise, and MCD-Nonlinear for nonlinear causal relationships and history-dependent noise. We demonstrate that our method surpasses state-of-the-art benchmarks in causal discovery tasks through extensive experimentation on synthetic and real-world datasets, particularly when the data emanates from diverse underlying causal graphs. Theoretically, we prove the identifiability of such a model under some mild assumptions. Implementation is available at https://***/Rose-STL-lab/MCD. Copyright © 2023, The Authors. All rights reserved.

关键词： Time series

来源：评论

学校读者我要写书评

暂无评论

In2NeCT: Inter-class and Intra-class Neural Collapse Tuning for Semantic Segmentation of Imbalanced Remote Sensing Images 39

In2NeCT: Inter-class and Intra-class Neural Collapse Tuning ...

引用

39th Annual AAAI Conference on Artificial Intelligence, AAAI 2025

作者： Shen, Junao Hu, Qiyun Feng, Tian Wang, Xinyu Cui, Hui Wu, Sensen Zhang, Wei School of Software Technology Zhejiang University China State Key Lab of CAD&CG Zhejiang University China Department of Computer Science and Information Technology La Trobe University Australia School of Earth Sciences Zhejiang University China Innovation Center of Yangtze River Delta Zhejiang University China

ISBN: (纸本)157735897X

Remote sensing images (RSIs) are frequently characterized by multi-scale inter-class objects and inconsistently distributed objects due to scene limitations, which would cause a significant data imbalance challenging the corresponding semantic segmentation. Recent methods have leveraged various deep learning techniques to capture high-quality representations for RSI semantic segmentation, but are hardly capable of addressing the afore-mentioned challenge given their limited explorations toward the mechanisms behind the representations. The recently discovered Neural Collapse (NC) phenomenon in computer vision models suggests the simplex equiangular tight frame (ETF) as the optimal representation structure, which has motivated us to observe that the optimal structure of last-layer representations is disrupted and inter-class representations for minor classes tend to become closer to each other beacuse of data imbalance. To address these issues, we propose Inter-class and Intra-class Neural Collapse Tuning (In2NeCT) to optimize the representations that satisfy the simplex ETF, which facilitates the discrimination of inter-class representations and the coherence of intra-class representations. Extensive experiments on three datasets demonstrate that our In2NeCT consistently leads to significant improvements in performance and outperforms the state-of-the-art methods. Copyright © 2025, Association for the Advancement of Artificial Intelligence (***). All rights reserved.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Spatio-Temporal-Frequency Graph Attention Convolutional Network for Aircraft Recognition Based on Heterogeneous Radar Network

arXiv

引用

arXiv 2022年

作者： Meng, Han Peng, Yuexing Wang, Wenbo Cheng, Peng Li, Yonghui Xiang, Wei School of Information and Communication Engineering Beijing University of Posts and Telecommunications Beijing100876 China Department of Computer Science and Information Technology La Trobe University Australia University of Sydney Australia School of Electrical and Information Engineering The University of Sydney NSW2006 Australia School of Engineering & Math Sciences La Trobe University Australia

This paper proposes a knowledge-and-data-driven graph neural network-based collaboration learning model for reliable aircraft recognition in a heterogeneous radar network. The aircraft recognizability analysis shows that: (1) the semantic feature of an aircraft is motion patterns driven by the kinetic characteristics, and (2) the grammatical features contained in the radar cross-section (RCS) signals present spatial-temporal-frequency (STF) diversity decided by both the electromagnetic radiation shape and motion pattern of the aircraft. Then a STF graph attention convolutional network (STFGACN) is developed to distill semantic features from the RCS signals received by the heterogeneous radar network. Extensive experiment results verify that the STFGACN outperforms the baseline methods in terms of detection accuracy, and ablation experiments are carried out to further show that the expansion of the information dimension can gain considerable benefits to perform robustly in the low signal-to-noise ratio region. Copyright © 2022, The Authors. All rights reserved.

关键词： Aircraft

来源：评论

学校读者我要写书评

暂无评论

End-to-end Hyperspectral Image Change Detection Network Based on Band Selection

arXiv

引用

arXiv 2023年

作者： Yao, Qingren Zhou, Yuan Tang, Chang Xiang, Wei School of Electrical and Information Engineering Tianjin University Tianjin300072 China School of Computer Science China University of Geosciences Wuhan430079 China School of Computing Engineering and Mathematical Sciences La Trobe University MelbourneVIC3086 Australia College of Science and Engineering James Cook University CairnsQLD4878 Australia

For hyperspectral image change detection (HSI-CD), one key challenge is to reduce band redundancy, as only a few bands are crucial for change detection while other bands may be adverse to it. However, most existing HSI-CD methods directly extract change feature from full-dimensional HSIs, suffering from a degradation of feature discrimination. To address this issue, we propose an end-to-end hyperspectral image change detection network with band selection (ECDBS), which effectively retains the critical bands to promote change detection. The main ingredients of the network are a deep learning based band selection module and cascading band-specific spatial attention (BSA) blocks. The band selection module can be seamlessly integrated with subsequent CD models for joint optimization and end-to-end reasoning, rather than as a step separate from change detection. The BSA block extracts features from each band using a tailored strategy. Unlike the typically used feature extraction strategy that uniformly processes all bands, the BSA blocks considers the differences in feature distributions among widely spaced bands, thereupon extracting more sufficient change feature. Experimental evaluations conducted on three widely used HSI-CD datasets demonstrate the effectiveness and superiority of our proposed method over other state-of-the-art techniques. Copyright © 2023, The Authors. All rights reserved.

关键词： Change detection

来源：评论

学校读者我要写书评

暂无评论

Vision and Olfactory-Based Wildfire Monitoring with Uncrewed Aircraft Systems

Vision and Olfactory-Based Wildfire Monitoring with Uncrewed...

引用

International Conference on Ubiquitous Robots (UR)

作者： Lingxiao Wang Shuo Pang Mantasha Noyela Kevin Adkins Lulu Sun Marwa El-Sayed Electrical Engineering Department Louisiana Tech University Ruston LA USA Electrical Engineering and Computer Science Department Embry-Riddle Aeronautical University Daytona Beach FL USA Computer Science Department Louisiana Tech University Ruston LA USA Aeronautical Science Department Embry-Riddle Aeronautical University Daytona Beach FL USA Engineering Fundamentals Department Embry-Riddle Aeronautical University Daytona Beach FL USA Civil Engineering Department Embry-Riddle Aeronautical University Daytona Beach FL USA

Wildfires threaten human lives, destroy facilities, and emit toxic smoke. Traditional wildfire monitoring methods are hindered by inflexibility (e.g., watch towers) and cannot provide precise geo-location of wildfires (e.g., satellites). Thanks to recent development in robotics, deploying uncrewed aircraft systems (UAS) to monitor wildfires has become a feasible solution. This article introduces a UAS-based wildfire monitoring system and implement it in a prescribed burn test. A multirotor UAS was employed as the search agent and carried both olfactory (i.e., carbon monoxide and particulate matter) and visual (i.e., a camera) sensors to detect the existence of wildfires. A fuzzy inference system is designed to fuse olfactory sensor outputs to estimate whether the UAS detects smoke. A deep learning model, i.e., You Only Look Once version 4 (YOLOv4), is employed to identify smoke from the captured images. We deployed the proposed UAS in a prescribed burn at Tallahassee, Florida, in May 2022. Experimental results show that the proposed fuzzy inference system improves the estimation accuracy of whether the UAS detects smoke compared with the fixed threshold algorithm. In addition, the proposed YOLOv4 model can also detect smoke from captured images with a small amount of training samples.

关键词：

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：