检索结果-内蒙古大学图书馆

A Virtual-Label-Based Hierarchical Domain Adaptation Method for Time-Series Classification

IEEE Transactions on Neural Networks and Learning Systems 2024年第6期36卷 11456-11465页

作者： Wenmian Yang Lizhi Cheng Mohamed Ragab Min Wu Sinno Jialin Pan Zhenghua Chen Institute of Artificial Intelligence and Future Networks Beijing Normal University (BNU) Zhuhai China China Telecom Cloud Technology Company Ltd. Shanghai China Technology and Research Institute for Infocomm Research (I2R) Agency for Science Fusionopolis Singapore Department of Computer Science and Engineering The Chinese University of Hong Kong (CUHK) Shatin NT Hong Kong

Unsupervised domain adaptation (UDA) is becoming a prominent solution for the domain-shift problem in many time-series classification tasks. With sequence properties, time-series data contain both local and sequential features, and the domain shift exists in both features. However, conventional UDA methods usually cannot distinguish those two features but mix them into one variable for direct alignment, which harms the performance. To address this problem, we propose a novel virtual-label-based hierarchical domain adaptation (VLH-DA) approach for time-series classification. Specifically, we first slice the original time-series data and introduce virtual labels to represent the type of each slice (called local patterns). With the help of virtual labels, we decompose the end-to-end (i.e., signal to time-series label) time-series task into two parts, i.e., signal sequence to local pattern sequence and local pattern sequence to time-series label. By decomposing the complex time-series UDA task into two simpler subtasks, the local features and sequential features can be aligned separately, making it easier to mitigate distribution discrepancies. Experiments on four public time-series datasets demonstrate that our VLH-DA outperforms all state-of-the-art (SOTA) methods.

关键词： Task analysis Feature extraction Training Natural language processing Human activity recognition Correlation Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Guiding Soft Robots with Motor-Imagery Brain Signals and Impedance Control

Guiding Soft Robots with Motor-Imagery Brain Signals and Imp...

引用

IEEE International Conference on Soft Robotics (RoboSoft)

作者： Maximilian Stölzle Sonal Santosh Baberwal Daniela Rus Shirley Coyle Cosimo Della Santina The Cognitive Robotics department Delft University of Technology Mekelweg 2 CD Delft Netherlands School of Electronics Engineering Dublin City University Dublin Ireland MIT Computer Science and Artificial Intelligence Laboratory (CSAIL) Massachusetts Institute of Technology Cambridge MA USA

Integrating Brain-Machine Interfaces into non-clinical applications like robot motion control remains difficult - despite remarkable advancements in clinical settings. Specifically, EEG-based motor imagery systems are still error-prone, posing safety risks when rigid robots operate near humans. This work presents an alternative pathway towards safe and effective operation by combining wearable EEG with physically embodied safety in soft robots. We introduce and test a pipeline that allows a user to move a soft robot's end effector in real time via brain waves that are measured by as few as three EEG channels. A robust motor imagery algorithm interprets the user's intentions to move the position of a virtual attractor to which the end effector is attracted, thanks to a new Cartesian impedance controller. We specifically focus here on planar soft robot-based architected metamaterials, which require the development of a novel control architecture to deal with the peculiar nonlinearities - e.g., non-affinity in control. We preliminarily but quantitatively evaluate the approach on the task of setpoint regulation. We observe that the user reaches the proximity of the setpoint in 66% of steps and that for successful steps, the average response time is 21.5s. We also demonstrate the execution of simple real-world tasks involving interaction with the environment, which would be extremely hard to perform if it were not for the robot's softness.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Toward a model-free feedback control of blood glucose induced by meals for type-1 diabetics: in silico studies

引用

Iran Journal of computer science 2024年第2期7卷 217-227页

作者： Selma, Boumediene Selma, Belkacem Abouaïssa, Hassane Chouraqui, Samira Kadem, Habib Mehdeb, Djamila Elaffifi, Mohamed Higher School of Agronomy Mostaganem 27000 Algeria Electrical Engineering Department Laboratory of Electrical Engineering and Renewable Energies LGEER Hassiba Benbouali University of Chlef B.P 78C Ouled Fares Chlef 02180 Algeria Univ.Artois UR 3926 Laboratoire de Génie Informatique et d’Automatique de l’Artois (LGI2A) Technoparc Futura Béthune F-62400 France Department of Computer Sciences University of Science and Technology of Oran Mohamed-Boudiaf USTOMB El Mnaouar BP 1505 Bir El Djir Oran 31000 Algeria Abdelhamid Ibn Badis University Mostaganem 27000 Algeria

This work addresses the problem of developing control algorithms for type 1 diabetic patients that provide an automatic connection between continuous glucose monitoring and insulin injection, without any manual intervention by the patient. This emerging technology is known as artificial pancreas (AP). In this study, a recent Model-Free Control (MFC) approach and its related “intelligent” regulators were exploited for this biomedical issue using an Extended Bergman Minimal Model. MFC is used to regulate the blood glucose (BG) level as well as to reduce postprandial glucose peak induced by meals in subjects with type 1 diabetes. Effects of unannounced meals three times a day are taken into account and are investigated. In silico results globally illustrate the feasibility of such a control. The proposed controller has proven its effectiveness in reducing postprandial BG rise, eliminating the effects of external disturbances such as the effect of meals, and maintaining a healthy BG level under uncertainties. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Artificial pancreas (AP) Blood glucose (BG) regulation Extended Bergman minimal model (EBMM) Model-free control (MFC) Type-1 diabetes (T1D)

来源：评论

学校读者我要写书评

暂无评论

Dog identification based on textural features and spatial relation of noseprint

引用

Pattern Recognition 2024年 151卷

作者： Chan, Yung-Kuan Lin, Chuen-Horng Wang, Ching-Lin Tu, Keng-Chang Yang, Shu-Chun Tsai, Meng-Hsiun Yu, Shyr-Shen Department of Management Information Systems National Chung Hsing University 145 Xingda Rd. South Dist. Taichung City402 Taiwan Department of Computer Science and Engineering National Chung Hsing University 145 Xingda Rd. South Dist. Taichung City402 Taiwan Department of Computer Science and Information Engineering National Taichung University of Science and Technology No. 129 Sec. 3 Sanmin Rd North Dist. Taichung City404 Taiwan Department of Information Management National Chin-Yi University of Technology No.57 Sec. 2 Zhongshan Rd. Taiping Dist. Taichung411030 Taiwan

This study proposes dog identification technology based on dog noseprints, which are equivalent to human fingerprints and possess unique characteristics. The aim is to utilize this technology for identifying and managing stray animals. The study presents three processing stages. In the first stage, YOLOv3 detects the dog's nose and nostril regions. The second stage involves enhancing the image's contrast and the contour of the scaly block using the multi-scale line detector. Finally, in the third stage, the shape and spatial features of the scaly block are extracted and utilized for dog identification. The study included a dataset of 276 dogs from multiple animal families and public shelters in Taiwan. The dataset was randomly divided into three groups to determine the optimal parameters for matching the dog noseprint via experimentation. Each dog identification group achieved an accuracy (ACC) exceeding 97.83 %, demonstrating that the proposed parameter-matching method offers high stability. Furthermore, in an additional experimental dataset consisting of dog noseprint images used for dog identification, the proposed method achieved an ACC exceeding 90.22 % in the Top 1 and 94.57 % in the Top 3. The ACC results across different groups consistently demonstrate the proposed method's ability to achieve high accuracy in dog identification. The source code and trained models are publicly available at: https://***/Chuen-HorngLin/Dog-Identification-Noseprint. © 2024 Elsevier Ltd

关键词： Animals

来源：评论

学校读者我要写书评

暂无评论

2-D GCANet Applied to Denoise 1-D EEG Signals in Online Remote Teaching Scenarios

Journal of Network Intelligence

引用

Journal of Network Intelligence 2023年第4期8卷 1289-1302页

作者： Huang, Ting-Qing Wang, Chuan-Sheng Chen, Zhao-Qi Zhang, Fu-Quan Meng, Xiang-Long Grau, Antoni Chen, Yang Huang, Jing-Wei College of Computer and Control Engineering Minjiang University Fuzhou University Town No. 200 Xiyuangong Road Fuzhou China Department of Automatic Control Technical Polytechnic University of Catalonia Autonomous Region of Catalonia Barcelona Spain College of Computer and Big Data Fuzhou University Fuzhou University Town No. 2 Wulong Jiangbei Avenue Fuzhou China Digital Media Art Key Laboratory of Sichuan Province Sichuan Conservatory of Music No. 2 Wannianchang Street Chenghua District Sichuan Province Chengdu City China Fuzhou Technology Innovation Center of intelligent Manufacturing information System Minjiang University Fuzhou University Town NO. 200 Xiyuangong Road Fuzhou China Fujian Province University No. 599 Quanxiu Road Juyuanzhou Ecological Civilization District Fujian Province Fuzhou City China College of Electronic Engineering Shandong University of Science and Technology No. 579 Qianwangang Road Huangdao District Qingdao China School of Mechanical and Automotive Engineering Fujian University of Technology No. 33 Xuefu South Road University New District Fujian Province Fuzhou City China

When EEG signals are used to assess the level of student engagement in online teaching tasks, they are often interfered by noise. It is a challenge to effectively remove these noises. Currently, deep learning methods have been applied to the field of EEG denoising. However, existing methods still have some problems. The denoised signals still have obvious noise residue, or the original EEG signals are damaged, and the model fitting speed is too slow. Image dehazing, as a typical denoising task in the field of image enhancement, has achieved great success in recent years. Therefore, inspired by advanced models in this field, we introduce CNN into EEG denoising tasks. In this paper, we take GCANet, an excellent image enhancement model, as an example. The dilated convolutions and gate fusion subnetworks included in GCANet enable more efficient EEG signal denoising. The results demonstrate that the proposed model effectively reduces noise while preserving essential features. Furthermore, in comparison to other state-of-the-art models, our proposed model exhibits enhanced robustness and faster convergence, as evidenced by achieving lower loss values after five epochs. Its good performance provides a new development idea for the field of EEG denoising. © 2023, Taiwan Ubiquitous Information CO LTD. All rights reserved.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Location Optimization and Resource Allocation of IRS in a Multi-User Indoor mmWave VR Network

Location Optimization and Resource Allocation of IRS in a Mu...

引用

IEEE Conference on Wireless Communications and Networking

作者： Jalal Jalali Maria Bustamante Madrid Filip Lemic Hina Tabassum Jakob Struye Jeroen Famaey Xavier Costa Perez Internet and Data Lab University of Antwerp - imec Belgium AI-Driven Systems Lab i2Cat Foundation Spain Electrical Engineering and Computer Science Department York University Canada NEC Labs Europe GmbH Germany ICREA Spain

ISBN: (数字)9798350303582

ISBN: (纸本)9798350303599

Next-generation Virtual Reality (VR) technology enables full-user immersion and support for multiuser Virtual Experiences (VEs). Given the low-cost and passive nature of intelligent reflecting surfaces (IRSs), this paper investigates the optimal design of a multi-user IRS-assisted VR network, where an IRS is optimally deployed in a confined space as a function of VR fully-immersed users' trajectory. In particular, we consider sum-rate maximization of all VR users and optimize the Access Point's (AP) active beamforming, and the IRS's placement, phase shifts, and radiation patterns in a confined indoor environment operating in millimeter Wave (mmWave) frequencies. We introduce the Alternating Optimization (AO) algorithm, decompose the problem into distinct sub-problems, and solve each problem optimally. That is, maximum-ratio transmission (MRT) is applied for optimal beamforming at the AP, optimal closed-from IRS phase shifts are determined using quadratic transformation, global optimization is conducted to determine the ideal locations for the IRS elements, and the monotonic optimal radiation pattern has been analyzed. Our findings highlight that strategically allocating the IRS's resources at optimal physical locations enhances signal stability and maximizes per-user throughput.

关键词： Wireless communication Surface waves Array signal processing Virtual reality Throughput Trajectory Resource management

来源：评论

学校读者我要写书评

暂无评论

Evolution equations on co-evolving graphs: long-time behaviour and the graph continuity equation

arXiv

引用

arXiv 2025年

作者： Carrillo, José Antonio Esposito, Antonio Mikolás, László Mathematical Institute University of Oxford Woodstock Road OxfordOX2 6GG United Kingdom Department of Information Engineering Computer Science and Mathematics Università degli Studi dell’Aquila Via Vetoio 1 Coppito L’Aquila67100 Italy

We focus on evolution equations on co-evolving, infinite, graphs and establish a rigorous link with a class of nonlinear continuity equations, whose vector fields depend on the graphs considered. More precisely, weak solutions of the so-called graph-continuity equation are shown to be the push-forward of their initial datum through the flow map solving the associated characteristics’ equation, which depends on the co-evolving graph considered. This connection can be used to prove contractions in a suitable distance, although the flow on the graphs requires a too limiting assumption on the overall flux. Therefore, we consider upwinding dynamics on graphs with pointwise and monotonic velocity and prove long-time convergence of the solutions towards the uniform mass *** Codes 35R02, 35A01, 35A02, 35A24, 35B40 Copyright © 2025, The Authors. All rights reserved.

关键词： Nonlinear equations

来源：评论

学校读者我要写书评

暂无评论

Quantum noise spectroscopy as an incoherent imaging problem

arXiv

引用

arXiv 2022年

作者： Tsang, Mankei Department of Electrical and Computer Engineering National University of Singapore 4 Engineering Drive 3 117583 Singapore Department of Physics National University of Singapore 2 Science Drive 3 117551 Singapore

I point out the mathematical correspondence between an incoherent imaging model proposed by my group in the study of quantum-inspired superresolution [Tsang, Nair, and Lu, Physical Review X 6, 031033 (2016)] and a noise spectroscopy model also proposed by us [Tsang and Nair, Physical Review A 86, 042115 (2012);Ng et al., Physical Review A 93, 042121 (2016)]. Both can be regarded as random displacement models, where the probability measure for the random displacement depends on unknown parameters. The spatial-mode demultiplexing (SPADE) method proposed for imaging is analogous to the spectral photon counting method proposed in Ng et al. (2016) for optical phase noise spectroscopy—Both methods are discrete-variable measurements that are superior to direct displacement measurements (direct imaging or homodyne detection) and can achieve the respective quantum limits. Inspired by SPADE, I propose a modification of spectral photon counting when the input field is squeezed—simply unsqueeze the output field before spectral photon counting. I show that this method is quantum-optimal and far superior to homodyne detection for both parameter estimation and detection, thus solving the open problems in Tsang and Nair (2012) and Ng et al. (2016). Copyright © 2022, The Authors. All rights reserved.

关键词： Homodyne detection

来源：评论

学校读者我要写书评

暂无评论

FruitsMultiNet: A deep neural network approach to identify fruits through multi-scale feature fusion using mobile interface

Journal of Agriculture and Food Research

引用

Journal of Agriculture and Food Research 2025年 22卷

作者： Tasauf Mim Md Mahbubur Rahman Jahanur Biswas Ahmad Shafkat Khandaker Mohammad Mohi Uddin Department of Computer Science and Engineering Jahangirnagar University Savar Dhaka 1342 Bangladesh Department of Computer Science and Engineering Dhaka International University Satarkul Rd Dhaka 1212 Bangladesh Department of Computer Science and Engineering Bangladesh University of Business and Technology Mirpur-2 Dhaka 1216 Bangladesh Department of Electrical Engineering and Computer Science (EECS) University of Wyoming Laramie WY 82071 USA Department of Computer Science and Engineering Southeast University Dhaka 1208 Bangladesh

The breakthrough of smart supply chain systems and the rapid adoption of automation in the growing fruit production and processing sectors are driving the increasing demand for an automatic fruit classification approach. A reliable fruit classification system during harvest and post-harvest phases can minimize time, cost, and human error while modernizing the processes of sorting, labeling, and packaging. The proposed research suggests a groundbreaking autonomous fruit classification method grounded on convolutional neural networks (CNNs). The classification of fruits is often challenging due to variations that occur in the same fruit throughout its life cycle. Additionally, the diverse compositions of sizes, shapes, and colors further complicate the process and impact accuracy. This research worked with a unique dataset of 3240 images of Bangladeshi fruits, publicly available on Kaggle. The data were preprocessed before the feature extraction phase to achieve more accurate outcomes. MobileNet, VGG16, NasNetMobile, DenseNet201, InceptionV3, and Xception were experimented with in the feature extraction and performance evaluation process. The proposed research combines MobileNet and VGG16 to build a transfer learning (TL) principles-based framework named FruitsMultiNet. These models were chosen because they achieved the highest individual accuracy compared to other models. The data were fragmented into training, testing, and validation sets for eight different fruits. The proposed FruitsMultiNet achieved a 99.84 % accuracy rate, which is greater than the accuracy rates of individual Deep Learning (DL) models. The integration of FruitsMultiNet into a mobile application makes it easily accessible for individuals, adding consistency and providing consumers with a more accessible system to classify fruits automatically. The application also provides nutritional information and health benefits of the recognized fruit for humans.

关键词： Deep learning FruitsMultiNet Fruits classification Convolution neural network Mobile interface

来源：评论

学校读者我要写书评

暂无评论

Correction of slanted text pictures based on modified opencv

Journal of Network Intelligence

引用

Journal of Network Intelligence 2021年第2期6卷 238-246页

作者： Lu, Ran Xu, Yongjin Wu, Zhanhe Zheng, Yijie Ren, Lingxin Ding, Feifei Li, Jianjun Chang, Chin-Chen State Grid Zhejiang Marketing Service Center Hangzhou311100 China School of Coomputer Science and Technology Hangzhou Dianzi University No.1 of the 2nd street XiaSha Hangzhou310018 China Department of Information and Engineering Computer Science Feng Chia University No. 100 Wenhwa Rd. Seatwen Taichung40724 Taiwan

Correction of slanted or tilted text is very important in many applications, such as optical character recognition (OCR), text detection and recognition. It is not difficult to find that the recognition accuracy of OCR has a great relationship with the tilt angle of the recognized picture, resulting in reduced OCR adaptability and great limitations. As for the problem, we proposed an innovative optimization to estimate the angle of tilting texts in a picture based on Hough transform and modified OpenCV, which reduces the negative effect of background noises. Our algorithm outperforms the state-of-the-art methods in both performances and robustness. © 2021 Global Research Online. All rights reserved.

关键词： Optical character recognition

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：