检索结果-内蒙古大学图书馆

Local Periodicity-Based Beat Tracking for Expressive Classical Piano Music

IEEE/ACM Transactions on Audio Speech and Language Processing 2023年 31卷 2824-2835页

作者： Chiu, Ching-Yu Muller, Meinard Davies, Matthew E. P. Su, Alvin Wen-Yu Yang, Yi-Hsuan National Cheng Kung University Academia Sinica Graduate Program of Multimedia Systems and Intelligent Computing Tainan701401 Taiwan International Audio Laboratories Erlangen Erlangen91058 Germany University of Coimbra Department of Informatics Engineering Centre for Informatics and Systems Coimbra3004-531 Portugal National Cheng Kung University Department of Computer Science and Information Engineering Tainan701401 Taiwan Taiwan AI Labs Yating Music Team Taipei103622 Taiwan Academia Sinica Research Center for IT Innovation Taipei11529 Taiwan

To model the periodicity of beats, state-of-the-art beat tracking systems use 'post-processing trackers' (PPTs) that rely on several empirically determined global assumptions for tempo transition, which work well for music with a steady tempo. For expressive classical music, however, these assumptions can be too rigid. With two large datasets of Western classical piano music, namely the Aligned Scores and Performances (ASAP) dataset and a dataset of Chopin's Mazurkas (Maz-5), we report on experiments showing the failure of existing PPTs to cope with local tempo changes, thus calling for new methods. In this paper, we propose a new local periodicity-based PPT, called predominant local pulse-based dynamic programming (PLPDP) tracking, that allows for more flexible tempo transitions. Specifically, the new PPT incorporates a method called 'predominant local pulses' (PLP) in combination with a dynamic programming (DP) component to jointly consider the locally detected periodicity and beat activation strength at each time instant. Accordingly, PLPDP accounts for the local periodicity, rather than relying on a global tempo assumption. Compared to existing PPTs, PLPDP particularly enhances the recall values at the cost of a lower precision, resulting in an overall improvement of F1-score for beat tracking in ASAP (from 0.473 to 0.493) and Maz-5 (from 0.595 to 0.838). © 2014 IEEE.

关键词： Hidden Markov models

来源：评论

学校读者我要写书评

暂无评论

Constructing an Automated Method for Interpreting Alzheimer’s Disease Based on Deep Learning Models and Image Processing 24

Constructing an Automated Method for Interpreting Alzheimer...

引用

8th International Conference on Medical and Health Informatics, ICMHI 2024

作者： Yi, Chia-Pin Ting, Hsien-Wei Yang, Tzu-Fang Chien, Ting-Ying Department of Computer Science and Engineering Yuan Ze University Taoyuan City Taiwan Department of Neurosurgery Taipei Hospital Ministry of Health and Welfare New Taipei City242 Taiwan Graduate Program in Biomedical Informatics Yuan Ze University Taoyuan City Taiwan TaiwanInnovation Center for Big Data and Digital Convergence Yuan Ze University Taoyuan City Taiwan

ISBN: (纸本)9798400716874

Alzheimer’s disease (AD) often presents only mild symptoms in its early stages, and as there is no direct diagnostic method currently available, many patients are diagnosed only after the condition has worsened. Considering that magnetic resonance imaging (MRI) can provide key information in the diagnosis of AD, but the condition may go undiagnosed under manual interpretation due to the subtleness of symptoms and the reliance on radiologists to interpret MRI images, this study aimed to develop an automated method for interpreting brain MR images, hoping to screen for potential AD patients. By combining image processing, the U-Net model, YOLOv8 model, various CNN model architectures, and ensemble data statistics, our method was able to identify features related to AD and predict whether a patient has the disease. Our method can process a complete set of MR images and automatically extract important slices for prediction, making it practical for medical applications. Furthermore, we tested the accuracy and reliability of the method with additional datasets. © 2024 Copyright ACM.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Two-Dimensional Positioning and Phase Synchronization System for Drone-Based Spatially Reconfigurable Phased Array

Two-Dimensional Positioning and Phase Synchronization System...

引用

IEEE Conference on Antenna Measurements & Applications (CAMA)

作者： Hsiang Cheng Sheng-Yeh Yang Tzu-Chun Su Yun-Ting Tsai Shih-Ming Huang Shih-Yuan Chen Graduate Institute of Communication Engineering National Taiwan University Taipei Taiwan Department of Electrical Engineering National Taiwan University Taipei Taiwan Department of Electrical and Computer Engineering University of California at Los Angeles California America

This paper presents a 2-D wireless ultrasound-microwave phase synchronization (UMPS) system for a Unmanned Aerial Vehicle (UAV) -based phased array. It is extended from its 1-D counterpart [6]. The UMPS system consists of multiple non-connected modules that operate in a leader-follower scheme for positioning and phase synchronization. The UMPS includes multiple non-connected modules that follow a leader-follower scheme for positioning and phase synchronization. The leader is the only reference point, to which followers locate their relative positions by Time-of-Flight (ToF) of ultrasound calibrated with microwave signal and synchronize its phase by on-board phase shifters. With a proper phase compensation, the UMPS modules can form a phased array with a 2D-variable spatial arrangement of antenna elements. In this work, the 2D-prototype UMPS exhibits a root-mean-square (RMS) positioning error at 9 mm and a 5.6° RMS phase error for synchronizing a 433 MHz continuous wave RF signal, which is satisfactory for forming a UAV-based phased array.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Accessing Gender Bias in Speech Processing Using Machine Learning and Deep Learning with Gender Balanced Audio Deepfake Dataset

Accessing Gender Bias in Speech Processing Using Machine Lea...

引用

International Conference on Informatics and Computing (ICIC)

作者： Tricia Estella Amalia Zahra Wai-Keung Fung Computer Science Department BINUS Graduate Program – Master of Computer Science Bina Nusantara University Jakarta Indonesia Department of Engineering Cardiff School of Technologies EUREKA Robotics Centre Cardiff Metropolitan University United Kingdom

ISBN: (数字)9798331517601

ISBN: (纸本)9798331517618

The rapid advancement of AI has led to the rise of Audio Deepfakes (AD), which pose serious ethical and security concerns by accurately mimicking human speech. This research addresses the urgent need for effective AD detection, with a focus on gender bias that can reduce the effectiveness of detection models. We examined how gender affects the performance of both Machine Learning (Support Vector Machine, Random Forest, Logistic Regression, XGBoost) and Deep Learning (Deep Neural Networks, Convolutional Neural Networks) models using the GBAD dataset. Our findings show that models trained on female audio outperform those trained on male audio, likely due to the expressive nature of female voice features and high-pitched artifacts in FAKE audio. This highlights the need for more robust, gender-sensitive detection systems. Future work should focus on developing adaptive models to reduce gender bias, improving security, and creating lightweight models for wider public use.

关键词： Deep learning Support vector machines Deepfakes Adaptation models Logistic regression Computational modeling Security Speech processing Informatics Random forests

来源：评论

学校读者我要写书评

暂无评论

3-D Positioning for Drone-Based Spatially Reconfigurable Phased Array

3-D Positioning for Drone-Based Spatially Reconfigurable Pha...

引用

IEEE Conference on Antenna Measurements & Applications (CAMA)

作者： Tzu-Chun Su Sheng-Yeh Yang Hsiang Cheng Yun-Ting Tsai Shih-Ming Huang Shih-Yuan Chen Department of Electrical Engineering National Taiwan University Taipei Taiwan Graduate Institute of Communication Engineering National Taiwan University Taipei Taiwan Department of Electrical and Computer Engineering University of California at Los Angeles Californica America

This study presents a 3D positioning method to enable precise phase synchronization among individual drones, which is the key for the drone-based spatially reconfigurable phased array [1]. The proposed system eliminates the dependency on GPS by enabling drones to locate each other as a cohesive unit. By employing a hierarchical structure consisting of leader and subleaders, coupled with an iterative method among drones, precise phase compensation is achieved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

cytoNet: Spatiotemporal network analysis of cell communities (vol 18, e1009846, 2022)

引用

PLOS COMPUTATIONAL BIOLOGY 2022年第11期18卷 e1009846页

作者： Mahadevan, A. S. Long, B. L. Hu, C. W. Ryan, D. T. Grandel, N. E. Department of Bioengineering University of Pennsylvania Philadelphia Pennsylvania United States of America. Department of Bioengineering Rice University Houston Texas United States of America. Systems Synthetic and Physical Biology Program Rice University Houston Texas United States of America. Department of Biomedical Engineering University of Texas at San Antonio San Antonio Texas United States of America. Biophysics Graduate Program University of California Berkeley California United States of America. Department of Oral & Maxillofacial Surgery University of Texas Health Science Center at San Antonio San Antonio Texas United States of America. Department of Electrical and Computer Engineering Rice University Houston Texas United States of America.

We introduce cytoNet, a cloud-based tool to characterize cell populations from microscopy images. cytoNet quantifies spatial topology and functional relationships in cell communities using principles of network science. Capturing multicellular dynamics through graph features, cytoNet also evaluates the effect of cell-cell interactions on individual cell phenotypes. We demonstrate cytoNet's capabilities in four case studies: 1) characterizing the temporal dynamics of neural progenitor cell communities during neural differentiation, 2) identifying communities of pain-sensing neurons in vivo, 3) capturing the effect of cell community on endothelial cell morphology, and 4) investigating the effect of laminin α4 on perivascular niches in adipose tissue. The analytical framework introduced here can be used to study the dynamics of complex cell communities in a quantitative manner, leading to a deeper understanding of environmental effects on cellular behavior. The versatile, cloud-based format of cytoNet makes the image analysis framework accessible to researchers across domains.

关键词： Calcium imaging Fluorescence imaging Growth factors Centrality Cell cycle and cell division Endothelial cells Network analysis Stem cells

来源：评论

学校读者我要写书评

暂无评论

Constrained Sorter Design using Zero-One Principle

Constrained Sorter Design using Zero-One Principle

引用

IEEE International Symposium on Circuits and Systems (ISCAS)

作者： Sangil Han Jaehee Kim Dongyun Kam Byeong Yong Kong Mijung Kim Young-Seok Kim Youngjoo Lee Department of Electrical Engineering POSTECH Pohang Korea Memory Business Samsung Electronics Hwaseong Korea Division of Electrical Electronic and Control Engineering Kongju National University Cheonan Korea Sports AIX Graduate Program POSTECH Pohang Korea

ISBN: (数字)9798350330991

ISBN: (纸本)9798350331004

To derive efficient sorting architectures constrained to application-specific input/output conditions, we present in this paper a systematic design methodology that can effectively prune dispensable compare-and-swap (CAS) units. Unlike the previous works resorting to heuristic approaches, the proposed framework exploits the zero-one principle to validate the pruning of a CAS unit at a time, generating the cost-optimized sorter architecture in an iterative manner with a reasonable complexity. In addition to the given input/output constraints, we newly develop the architecture options for the proposed framework, allowing more design spaces for finding the most attractive constrained-sorter design. For 8-list polar decoders, the proposed framework successfully reduces 70% of CAS units in the baseline full sorter, relaxing the area-time complexity by 35% compared with the state-of-the-art solutions.

关键词： Systematics Circuits and systems Design methodology computer architecture Hardware Complexity theory Decoding

来源：评论

学校读者我要写书评

暂无评论

An Intrusion Detection System for IoT Using KNN and Decision-Tree Based Classification

An Intrusion Detection System for IoT Using KNN and Decision...

引用

2021 International Conference of Modern Trends in Information and Communication Technology Industry, MTICTI 2021

作者： Abdaljabar, Zainab Hussam Ucan, Osman Nuri Ali Alheeti, Khattab M. University of Altinbaş Institute of Graduate Studies Department of Electrical and Computer Engineering Turkey University of Anbar College of Computer Science and Information Technology Department of Computer Networking Systems Iraq

ISBN: (纸本)9781665414449

The Internet of Things (IoT) has grown rapidly in recent years, intending to affect everything from everyday life to large industrial systems. Regrettably, this has attracted the attention of hackers, who have turned the Internet of Things into a target for malicious activity, exposing end nodes to attack. IoT devices' sheer volume and diversity make protecting the IoT infrastructure with a traditional intrusion detection system difficult. So to protect IoT devices, the data flow was investigated in an IoT context to protect these devices from hackers. We used two machine learning classifiers in this work: KNN (K-Nearest Neighbors) and DT (Decision Tree). We calculated the Error Rate, Accuracy, Precision, Recall, and F1 score for each method. When we combined these two classifiers, we obtained outstanding results (100 %). We have a high rate of detection of attacks. The findings are summarized. © 2021 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Purchase intention of Amazon sustainable brands: a hybrid approach of structural equation modeling and Rasch model

引用

SN Business and Economics 2023年第12期3卷 1-29页

作者： Santos, Luiz Diego Vidal Holanda, Francisco Sandro Rodrigues de Santana Oliveira, Catuxe Varjão Pedrotti, Alceu de Menezes, Pedro Vinícius Bertulino Villwock, Ana Paula Schervinski Graduate Program in Intellectual Property Science Universidade Federal de Sergipe-UFS Sergipe São Cristóvão Brazil Agronomy Engineering Department Universidade Federal de Sergipe-UFS Sergipe São Cristóvão Brazil Computer Science Department Instituto Federal de Sergipe-UFS Sergipe Lagarto Brazil

This article conducts a comprehensive examination of brand-based equity closely linked to environmental sustainability in the Amazon Rainforest. The research is primarily dedicated to identifying the factors that influence consumers’ willingness to purchase products from sustainable brands within six categories: cosmetics, tourism, clothing, accessories, and food. The central aim of this study is to investigate how brand equity constructs, rooted in environmental sustainability within the Amazon Rainforest, impact consumer loyalty and purchase intention. To accomplish this objective, a mixed-methodology approach was employed, combining structural equation modeling with item response theory. This study involved 535 potential consumers, and data from eight selected brands served as the basis for the analysis. Within the framework of Purchase Intention (PI), the following variables were regarded as formative constructs: Strategic Brand Positioning (SBP), Perceived Brand Quality (PBQ), Retail Innovation (RI), and Brand Authenticity (BA). The results obtained from the structural equation models underscore the statistical significance and substantial impact of all pathways. Notably, SBP emerged as the most significant contributor to the effect on PI, as hypothesized. Brand equity and environmental sustainability were found to indirectly affect behavioral PI. Consumers exhibited a high level of latent trait (θ), with items consistently reflecting elevated values, confirming their acceptance of products, except brand differentiation. Overall, the findings highlight the effectiveness of branding strategies rooted in environmental sustainability in enhancing the brand equity of small forest companies. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2023.

关键词： Amazon Rainforest Brands Entrepreneurship Microenterprises Sustainability performance Sustainable development

来源：评论

学校读者我要写书评

暂无评论

Exploring a Human-in-the-Loop Framework for Adaptive Sign Language Translation in Deaf Education

Exploring a Human-in-the-Loop Framework for Adaptive Sign La...

引用

International Conference on engineering of Modern Electric Systems (EMES)

作者： I Dewa Made Bayu Atmaja Darmawan Linawati Gede Sukadarmika Ni Made Ary Esta Dewi Wirastuti Reza Pulungan Dewa Putu Yudhi Ardiana Doctoral Engineering Department Udayana University Denpasar Indonesia Electrical Engineering Department Udayana University Denpasar Indonesia Computer Science and Electronics Department Universitas Gadjah Mada Yogyakarta Indonesia Graduate School of Information Science and Engineering Ritsumeikan University Osaka Japan

ISBN: (数字)9798331525774

ISBN: (纸本)9798331525781

Sign Language Translation (SLT) plays a crucial role in enhancing accessibility and literacy for the deaf community. However, SLT models face significant challenges, including limited annotated datasets, sign language variability, and generalization issues. This study proposes a Human-in-theLoop (HITL) SLT conceptual framework to enhance model adaptability through continuous teacher and student interactions. By integrating Federated Learning (FL), schools can train SLT models locally while preserving data privacy and accommodating regional sign language variations. The proposed framework enables teachers to refine model predictions via Active Learning, introduce new sign gestures using Few-shot Learning, and enhance adaptability through Reinforcement Learning. Additionally, FL facilitates decentralized model updates, reducing computational burdens while improving model inclusivity across multiple schools. The synergy between HITL and FL creates an adaptive system that evolves based on real-time feedback, making AI-driven SLT more effective in educational settings. Future directions include deploying the HITL-SLT model in real-world classrooms to evaluate its practical impact. Conducting small-scale studies with teachers and students will be essential for validating its effectiveness, refining model adaptability, and enhancing interactive learning experiences.

关键词： Adaptation models Sign language Data privacy Translation Federated learning Computational modeling Reinforcement learning Human in the loop Real-time systems Few shot learning

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：