检索结果-内蒙古大学图书馆

Anomalous Sound Detection Using Time-Frequency Feature and Mixbatch

Journal of Shanghai Jiaotong University (science) 2025年 1-8页

作者： Huang, Shun Zhang, Yunxiang Fang, Zhihua Tang, Minrui Xu, Ruifeng He, Liang School of Computer Science and Technology Xinjiang University Urumqi830017 China Xinjiang Key Laboratory of Signal Detection and Processing Urumqi830017 China School of Intelligence Science and Technology Xinjiang University Urumqi830017 China Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Tsinghua University Beijing100084 China

The sound emitted by machines under abnormal working conditions exhibits various frequency patterns. Currently, the most advanced anomalous sound detection (ASD) approach is to apply a multi-head self-attention mechanism to the Log-Mel spectrogram for automatic frequency pattern analysis. However, the Log-Mel spectrogram may filter out high-frequency components of abnormal sounds;thus the use of self-attention mechanisms on Log-Mel spectrogram seems to have certain limitations. In this paper, we construct a simple convolutional neural network to extract comprehensive frequency features from raw audio to complement spectral-temporal information fusion. The parameters of this neural network are continuously updated during the training process to extract better frequency features for downstream classification neural networks in ASD. Additionally, a method for data augmentation in the batch dimension was developed to help the classification model learn both types of comprehensive features simultaneously. The proposed method achieved an AUC of 94.13% and a pAUC of 89.09% on the DCASE 2020 Challenge Task 2 dataset. Even when using only the features proposed in this paper, an AUC of 84.62% was achieved. (Figure presented.). © Shanghai Jiao Tong University 2025.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

Resource Allocation for Fairness Enhancement in Multi-Cell Vehicular VLC System With Optical IRS: A Cooperative Transmission Approach

引用

IEEE Internet of Things Journal 2025年第11期12卷 16931-16946页

作者： An, Nan Yang, Fang Liu, Chang Cheng, Ling Song, Jian Han, Zhu Tsinghua University Beijing National Research Center for Information Science and Technology Department of Electronic Engineering Beijing100084 China Center of National Railway Intelligent Transportation System Engineering and Technology Beijing100081 China China Academy of Railway Sciences Corporation Limited Signal and Communication Research Institute Beijing100081 China University of the Witwatersrand School of Electrical and Information Engineering Johannesburg2050 South Africa Tsinghua University Department of Electronic Engineering Beijing National Research Center for Information Science and Technology Beijing100084 China Tsinghua University Shenzhen International Graduate School Shenzhen518055 China University of Houston Department of Electrical and Computer Engineering HoustonTX77004 United States Kyung Hee University Department of Computer Science and Engineering Seoul446-701 Korea Republic of

In multi-cell downlink vehicular visible light communication (VLC) systems, vehicles at a cell edge experience lower achievable data rates compared with those at a cell center, primarily due to the weaker channel gain and inter-cell interference, leading to unfairness among vehicles. In this paper, a cooperative transmission approach is proposed for a vehicular VLC system with optical intelligent reflecting surface (OIRS) to enhance the max-min fairness of the system by leveraging the additional OIRS-reflected channels and the interference-mitigating capabilities of cooperative transmission. To this end, the system model is established, followed by the formulation of a resource allocation problem aiming at enhancing max-min fairness, in which the minimum achievable data rate among vehicles is optimized. Then, an effective resource algorithm is proposed, transforming the original problem into an equivalent form and subsequently decomposing it into three subproblems, focusing on OIRS assignment, subchannel allocation, and power adjustment, respectively. By employing a block coordinate descent algorithm, the three subproblems are solved iteratively until convergence. In addition, simulation results confirm the improvement in max-min fairness brought by OIRS and the proposed cooperative transmission approach. Moreover, compared with various baselines, the proposed resource allocation algorithm significantly enhances in the max-min fairness for the vehicular VLC system, crucial for this application. © 2014 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Integrating Spectro-Temporal Cross Aggregation and Multi-Scale Dynamic Learning for Audio Deepfake Detection

Integrating Spectro-Temporal Cross Aggregation and Multi-Sca...

引用

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

作者： Hao, Yunqi Xu, Minqiang Chen, Yihao Liu, Yanyan He, Liang Fang, Lei Liu, Lin School of Computer Science and Technology Xinjiang University Urumqi China Hefei iFly Digital Technology Co. Ltd. Hefei China University of Science and Technology of China Hefei China Department of Electronic Engineering Tsinghua University Beijing China

ISBN: (纸本)9798350368741

Audio deepfake refers to the technology of synthesizing speech using deep learning or large model algorithms. Compared to human voice, synthetic deepfake speech exhibits artifacts at global and local levels, which can be leveraged by audio deepfake detection (ADD) to distinguish real and fake speech. In this paper, we designed the spectro-temporal cross aggregation (STCA) module and the local multi-scale dynamic convolution (LMDC) module to extract global and local artifacts for detecting forged information, respectively. The STCA module utilizes a dual-branch structure with cross-attention, extracting global temporal and frequency features through the branches and aggregating mutual artifacts via cross-attention. The LMDC module uses multi-scale dynamic convolution for grouped features, to extract local information. Experimental results on multiple test sets demonstrate the effectiveness of our method. Specifically, we achieved an EER of 1.87% on the ASVspoof2021 DF evaluation set, surpassing the current state-of-the-art system by a relative 14.6%. © 2025 IEEE.

关键词： Audio deepfake detection Global-local modeling Multi-scale feature XLS-R

来源：评论

学校读者我要写书评

暂无评论

Morphable 3D architectures enabled by shear-guided approach

引用

Materials Today 2025年 86卷 28-41页

作者： Li, Kan Liu, Zhi Hu, Xiaonan Xu, Zijian Zhou, Yunlei Chen, Lin Huang, Yonggang Bo, Renheng Zhang, Yihui AML Department of Engineering Mechanics State Key Laboratory of Flexible Electronics Technology Tsinghua University Beijing100084 China State Key Laboratory of Digital Manufacture Equipment and Technology Huazhong University of Science and Technology Wuhan430074 China State Key Laboratory for Mechanical Behavior of Materials School of Materials Science and Engineering Xi'an Jiaotong University Shaanxi Xi'an710000 China Departments of Mechanical Engineering Civil and Environmental Engineering Materials Science and Engineering and Center for Bio-Integrated Electronics Northwestern University EvanstonIL60208 United States

3D structures and electronic devices assembled through mechanically guided approaches have found important applications in a wide spectrum of emerging fields, including healthcare, biomedical engineering, wearable devices, robotics, among others. To broaden the obtainable functional 3D architectures via mechanical pathways, in this work, a shear-guided assembly approach is demonstrated. A series of structures in forms of ribbon, membrane and hybrid configurations with previously inaccessible geometrical features, such as reversible local flips and centrosymmetric charity are showcased. Finite element analysis (FEA) is performed, suggesting that such shear-induced structures are capable of reversible post-assembly shape transformation among different modes. Given the above, a facile design, consisting of two ribbon-like 3D units (dual-3D-unit design), is introduced and carefully calibrated for shear sensing. Notably, such design is capable of translating the hard-to-measure shear deformation into easily-accessed vertical displacements while insensitive to finite tensile loadings. A multifunctional arrayed electronic interface is fabricated using such dual-3D-unit design, suggesting promising potentials in human–machine interactions. © 2025 Elsevier Ltd

关键词： Shear deformation

来源：评论

学校读者我要写书评

暂无评论

AtheroRisk: An Integrated computer Software System for Stroke Risk Stratification Utilizing Carotid Plaque Analysis in Ultrasound Videos

引用

SN computer science 2025年第2期6.0卷 1-16页

作者： Gemenaris, Michalis Liapi, Georgia D. Markides, Christos Constantinou, Kyriacos Loizou, Christos P. Neophytou, Michalis Kardoulas, Dimitrios Pattichis, Marios S. Griffin, Maura Nicolaides, Andrew Pattichis, Constantinos S. Kyriacou, Efthyvoulos Department of Electrical Engineering Computer Engineering and Informatics Cyprus University of Technology Limassol Cyprus Department of Computer Science and Engineering Frederick University Nicosia Cyprus Department of Computer Science University of Cyprus Nicosia Cyprus American Medical Center Nicosia Cyprus Vascular Surgery Clinic Attikon General University Hospital Athens Greece Department of Electrical and Comp. Engineering Univ. of New Mexico Albuquerque NM United States Vascular Screening and Diagnostic Center Nicosia Cyprus

In this study we propose and evaluate AtheroRisk, a standalone integrated computer software system for the analysis of carotid B-mode ultrasound (U/S) images and videos. Our goal was to provide a tool to help physicians stratify stroke risk. The presented system is based on outcomes from different research studies, as well as European guidelines for the evaluation and treatment of carotid artery disease. The goal is to enable reproducible analysis of atherosclerotic plaques in real time. AtheroRisk brings together analysis of U/S image and/or video of the carotid arteries facilitating anonymization, standardization, noise filtering and segmentation of plaques;followed by image and motion feature extraction to derive plaque-composition and stability. All the above steps are combined to determine the annual stroke risk rate and the 5-year stroke-free survival rate. The system is powered by an SQLite local database, in which the end user can save and manage data extracted from processing and analysis in an anonymized way. The first version of AtheroRisk was developed following an incremental software development process. The developed system was verified using 54 U/S videos (27 Asymptomatic, AS;27 Symptomatic, SY cases) while the clinician’s satisfaction and comments were collected through a questionnaire-based validation process. The results show that the integrated system proposed in this study can be successfully used for the automated image and video analysis of the CCA plaque in ultrasound videos. © The Author(s) 2025.

关键词： Carotid plaque analysis Motion analysis Plaque features Stroke risk stratification Ultrasound video

来源：评论

学校读者我要写书评

暂无评论

UltraSegNet:A Hybrid Deep Learning Framework for Enhanced Breast Cancer Segmentation and Classification on Ultrasound Images

引用

computers, Materials & Continua 2025年第5期83卷 3303-3333页

作者： Suhaila Abuowaida Hamza Abu Owida Deema Mohammed Alsekait Nawaf Alshdaifat Diaa Salama Abd Elminaam Mohammad Alshinwan Department of Computer Science Faculty of Prince Al-Hussein Bin Abdallah II for Information TechnologyAl Al-Bayt UniversityMafraq25113Jordan Medical Engineering Department Faculty of EngineeringAl-Ahliyya Amman UniversityAmman19328Jordan Department of Information Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Faculty of Information Technology Applied Science Private UniversityAmman11931Jordan Faculty of Computers Science Misr International UniversityCairo11800Egypt Jadara Research Center Jadara UniversityIrbid21110Jordan

Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image *** paper presents the UltraSegNet architecture that addresses these challenges through three key technical innovations:This work adds three things:(1)a changed ResNet-50 backbone with sequential 3×3 convolutions to keep fine anatomical details that are needed for finding lesion boundaries;(2)a computationally efficient regional attention mechanism that works on high-resolution features without using a transformer’s extra memory;and(3)an adaptive feature fusion strategy that changes local and global featuresbasedonhowthe image isbeing *** evaluation on two distinct datasets demonstrates UltraSegNet’s superior performance:On the BUSI dataset,it obtains a precision of 0.915,a recall of 0.908,and an F1 score of *** the UDAIT dataset,it achieves robust performance across the board,with a precision of 0.901 and recall of ***,these improvements are achieved at clinically feasible computation times,taking 235 ms per image on standard GPU ***,UltraSegNet does amazingly well on difficult small lesions(less than 10 mm),achieving a detection accuracy of *** is a huge improvement over traditional methods that have a hard time with small-scale features,as standard models can only achieve 0.63–0.71 *** improvement in small lesion detection is particularly crucial for early-stage breast cancer *** from this work demonstrate that UltraSegNet can be practically deployable in clinical workflows to improve breast cancer screening accuracy.

关键词： Breast cancer ultrasound image segmentation classification deep learning

来源：评论

学校读者我要写书评

暂无评论

A Systematic Flexible-Window based Scheduling Framework for Time-Sensitive Networking

引用

IEEE Internet of Things Journal 2025年第11期12卷 15636-15649页

作者： Sun, Wenjing Zou, Yuan Guan, Nan Zhang, Xudong Liu, Jiahui Farzaneh, Morteza Hashemi Beijing Institute of Technology National Engineering Laboratory for Electric Vehicles School of Mechanical Engineering and Collaborative Innovation Center of Electric Vehicles in Beijing Beijing100081 China City University of Hong Kong Department of Computer Science Hong Kong RealtimeIT Germany

Time-Sensitive Networking (TSN) is increasingly applied in automotive and industrial internet fields due to its low latency and deterministic communication. Gate control list (GCL) is foundational for deploying TSN. Currently, most scheduling research focuses on frame-to-window based scheduling. This scheduling approach typically generates a specific window for each frame, leading to a proliferation of GCL in large networks, which increases the complexity of implementing TSN. To simplify deployment and enhance scheduling reliability, this paper introduces a systematic flexible-window based scheduling framework. Utilizing a gapless GCL design approach, it optimizes flow's worst-case end-to-end (e2e) delays through window length design, with delays obtained through network calculus analysis. A generic solving framework based on metaheuristic algorithms is established to address this optimization problem. The scheduling framework also features a load-balanced turn prohibition routing strategy to balance link loads and avoid cyclic dependencies, alongside a K-means priority clustering method based on routing overlap to reduce the number of priorities. Simulation validation in a high-level autonomous driving vehicle's in-vehicle network shows that the proposed method can decrease GCL numbers by nearly 90% against frame-to-window scheduling. In common industrial internet scenario, it significantly reduces worst-case e2e delays and enhances scheduling success rates compared to the analogous scheduling method. Large-scale complex network scenario further demonstrates its scalability. © 2025 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Integrating multi-modal insights with transfer learning for detecting metastatic breast cancer (MBC-stage IV) prognostics

引用

International Journal of Information technology (Singapore) 2025年第1期17卷 637-643页

作者： Kaur, Sugandha Kaur, Manpreet Khanna, Ashish Department of Computer Science Mata Sundri College for Women University of Delhi New Delhi India Department of Computer Science & Technology Manav Rachna University Faridabad India Department of Computer Science and Engineering Maharaja Agrasen Institute of Technology New Delhi India Center for Global Health Research Saveetha Medical College Saveetha Institute of Medical and Technical Sciences Chennai India

Accurate breast cancer prognosis prediction is crucial for efficient treatment planning and improving patient outcomes. While significant progress has been made in treating primary breast cancer, the development of robust predictive models remains a critical challenge. This study introduces multi-modal prognostic approach for enhanced Metastatic Breast Cancer prediction. Recognizing the limitations of relying solely on uni-modal data, our approach leverages the power of multi-modal datasets. The images are taken from breast ultrasound scans dataset and mammograms are resized into 224 × 224, and finally, data pre-processing steps are used. This study presents a two-stage model: first, a convolutional neural network is used to extract salient features from the multi-modal data. These extracted features are further concatenated for feature-fusion and served as an input for a dual transfer learning hybrid model in the second stage by combining EfficientNetB4 and InceptionV3, thus enabling robust and accurate prognosis prediction. The outputs are flattened and concatenated followed by implementation of Deep_dense layers to learn the weights of the combined models. The proposed hybrid approach effectively categorizes medical images for the prognostics of breast-cancer. Evaluations demonstrate the superior performance of our proposed model, achieving an accuracy of 99.12%, precision of 95%, recall of 93%, and F1 score of 98.12%. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Convolutional neural network (CNN) EfficientNetB Feature fusion InceptionV3 Metastatic breast-cancer Multi-modal image data

来源：评论

学校读者我要写书评

暂无评论

HPPH: computer Vision-Based Service for High-Performance Pavement Health Recognition

引用

IEEE Internet of Things Journal 2025年第11期12卷 15987-15996页

作者： Tsung, Chen-Kun Kristiani, Endah Chiu, Chen-Kang Liu, Jung-Chun Yang, Chao-Tung National Chin-Yi University of Technology Department of Computer Science and Information Engineering Taichung City411030 Taiwan Tunghai University Department of Computer Science Taichung City407224 Taiwan Krida Wacana Christian University Department of Informatics Jakarta11470 Indonesia Tunghai University Research Center of Smart Sustainable Circular Economy Taichung City407224 Taiwan

Identifying pavement damage is a crucial component in road maintenance and infrastructure management. Prompt detection and corrective action of pavement defects can prevent severe deterioration, maintain safety, and prolong the useful life of road infrastructure. computer vision is widely applied to vehicle applications, such as driver assistance and self-driving, and that is a good platform to equip the pavement damage recognition service. Therefore, this work applies computer vision to develop the pavement damage detection service named high-performance pavement health recognition (HPPH) to detect common pavement damages, including longitudinal cracks, transverse cracks, alligator cracks, and potholes. The proposed HPPH can be deployed on a vehicle for regular road detection. To fit the requirements of vehicle applications, HPPH considers state-of-the-art techniques to optimize the performance of the recognition service on vehicles, e.g., DeepStream is applied to increase the inference performance. At the same time, YOLOv8s (You Only Look Once, YOLO) provides real-time inference. The experimental results reveal that HPPH provides an accuracy of 90.3% and an average precision 0.5 of 74.3%. Moreover, the optimized HPPH provides three times better than the pure YOLOv8s in terms of the inference speed. In summary, the proposed HPPH provides high recognition accuracy with high efficiency and is feasible to be applied to vehicle applications to realize distributed road maintenance. © 2025 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Improving Food Safety by IoT-based Climate Monitoring and Control Systems for Food Processing Plants

引用

Remote Sensing in Earth Systems sciences 2025年第2期8卷 387-399页

作者： Kolikipogu, Ramakrishna Shivaputra Muniyandy, Elangovan Maroor, Jnaneshwar Pai Lakshmi, Gundala Venkata Rama Konduri, Bhagavan Naveenkumar, R. Department of Information Technology Chaitanya Bharathi Institute of Technology Hyderabad India Department of Electronics and Communication Engineering Dr. Ambedkar Institute of Technology Bengaluru India Department of Biosciences Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Chennai India Applied Science Research Center Applied Science Private University Amman Jordan Department of Business Administration Nitte Institute of Professional Education Nitte (Deemed to Be University) Karnataka Mangaluru India Department of Computer Science and Engineering CVR College of Engineering Hyderabad India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Vaddeswaram Andhra Pradesh India Department of Computer Science and Engineering Chandigarh Colleges of Engineering Chandigarh Group of Colleges Jhanjeri Punjab Mohali 140307 India

Web different IoT apps offer lot more options for every player in the meat supply chain to monitor, control, and optimize their manufacturing method. Beginning within late 2017, industrial IoF2020 program will concentrate toward massive deployments of IoT via three use cases in the meat supply chain: pig farm management, poultry chain tracking, and for livestock visibility with openness. This initiative will include eleven collaborators spanning 5 nations. Installing and integrating IoT sensors for environmental and livestock management would alleviate producers’ difficulty getting knowledge that data needed regarding continuously tracking how well they are doing. Alert detection is additionally being created, integrating multiple sources of data in order to give farmers helpful input along with communicating news with other partners. By carrying out this, preventative or remedial measures can be done for infections, swine taint, bird mortality, feed waste, the natural world, et cetera. Additionally, EPCIS-based traceability from farm to fork will be allowed to ensure customers may get precise data about the source and caliber of their livestock. Each of these three use scenarios’ existing stage in work along with future plans are expected to be discussed. IoF2020 seeks to support the digital revolution in smart farming by tackling various educational and commercial problems while also promoting its use throughout Europe. It maintains mechanical necessary degree and security against physical along with toxic contaminants during industrial butchering of chicken, a lot measures have been taken. Buyers’ worries about food safety and health are the result of eating chicken increasingly. Our authorities have begun many kinds of initiatives aimed at improving safeguards for food. Our goal is to create an apparatus that alerts the factory farms affecting potential threats for product hygiene of food. With such setup, the detector can be mounted on top of the poultry slaugh

关键词： Food safety Internet of Things Meat processing Monitoring Supply chain

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：