检索结果-内蒙古大学图书馆

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

作者： Paul, Topon reddy, Vidhisha Ayyagari, Sai Prem Kumar Shingaki, ryusei Nishino, Kaneharu Corporate R&d Center Toshiba Corporation System Ai Laboratory Kanagawa212-8582 Japan Pvt. Ltd. R&d Division KA Bangalore560034 India

ISBN: (纸本)9798350374889

Low visibility condition affects the safety and operation of road, sea, and air traffic, which causes disruption in transportation and sometimes leads to serious accidents. Therefore, forecasting of low visibility accurately is of utmost importance for transportation agencies and aviation authorities. The conventional approach of visibility forecasting using time-series weather data cannot forecast low visibility with high accuracy due to imbalanced and clipped visibility data used to build a forecasting model. To address this issue, we propose a new approach that combines forecasting and classification methods with discrete and kernel density estimation (KdE)-based weights to build forecasting models for multi-step ahead forecasting. We show the effectiveness of our proposed approach by performing experiments with four publicly available visibility datasets of weather and air quality data from the USA and Japan and evaluating in terms of forecasting errors and accuracy of forecasting of low visibility. Experimental results suggest that our proposed approach can improve rMSE and MAE in the low visibility range by respectively 29-54 % and 40-59 % and F-score of forecasting of low and high visibility by 0.20-0.26 on the average over existing technologies for 3-24 hour ahead forecasting, which implies that it is well-suited for forecasting of low visibility. © 2024 IEEE.

关键词： Wind forecasting

来源：评论

学校读者我要写书评

暂无评论

research on Spider Fine-Grained recognition Technology Based on Transfer Learning

引用

Journal of Computing Science and Engineering 2023年第4期17卷 145-160页

作者： Wang, Jianming deng, Longfeng Shi, Chenyang Ye, Guosheng Yang, Zizhong School of Mathematics and Computer Science Dali University Dali China Yunnan Provincial Key Laboratory of Entomological Biopharmaceutical R&D Dali University Dali China

Few-shot image recognition represents a critical challenge in computer vision research. The scarcity of samples often results in inaccurate classification, limited generalization capabilities, and overfitted model recognition. To address these issues, the present study focuses on spider image recognition utilizing transfer learning and data augmentation techniques in limited sample settings. First, the BasNet image segmentation model and background replacement algorithm are used to extract species image data from the foreground;data augmentation is then applied to address the scarcity of samples. Second, a layer-by-layer fine-tuned transfer learning strategy based on the resNet-50 model is devised. Specifically, to mitigate overfitting in the few-shot image classification task, the first two residual blocks are frozen so that only the last two are trained. To enhance the model’s representation and generalization abilities, the SSC-resNet-50 optimization model is constructed by introducing symmetry techniques. This study aims to enhance the accuracy and performance of spider image recognition. The experimental results demonstrate that the improved SSC-resNet-50 model achieves an average accuracy of 99.1% in recognizing five types of spiders, thereby surpassing the performance of traditional models. These findings offer valuable insights for the field of small-sample high-precision image recognition. Category: computer Graphics/Image Processing © 2023, This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://***/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

radar Based Gesture recognition in Embedded systems 6

Radar Based Gesture Recognition in Embedded Systems

引用

6th International Conference on Electronics Technology, ICET 2023

作者： Liang, Tingxuan Lin, Yue Xu, Hongtao Fudan University State Key Laboratory of Integrated Chips & System Department of Microelectronics Shanghai China R&d ICLegend Micro Nanjing China

ISBN: (纸本)9798350337693

Gesture recognition has played a significant role in human-computer interaction. Millimeter-wave radar sensor is a class of necessary sensor for human gesture recognition in an indoor condition due to its advantages of private protection, low hardware cost. However, for gesture recognition application, it is difficult for us to meet the requirements of high accuracy and real-time simultaneously in embedded system. Unlike other conventional method which increase the cost of hardware of embedded system to improve detection accuracy, in this paper, we proposed a radar-based gesture recognition in low power consumption and low-cost embedded system by using 24 GHz millimeter-wave radar with 1 transmitting antenna and 2 receiving antenna channels. Besides, 2d trajectory information on XY coordination and lightweight deep neural network have been used to distinct ten air-writing numbers. As a result, this system could reconstruct the air-writing symbols and recognize numeric character 0-9 with 95.2% accuracy in real time. © 2023 IEEE.

关键词： Embedded systems

来源：评论

学校读者我要写书评

暂无评论

Solve the False-Positive Problem of the Mimic system Based on the Best Mimic Component Set

引用

China Communications 2022年第5期19卷 253-266页

作者： Yuwen Shao Zheng Zhang Xiaomei Wang Chuanxing Pan Jiangxing Wu State Key Laboratory of Mathematical Engineering and Advanced Computing Zhengzhou 450001China China National Digital Switching System Engineering and Technological R&D Center Zhengzhou 450002China

At present,there is a problem of false positives caused by the too vast mimic scope in mimic transformation *** studies have focused on the“compensation”method to deal with this problem,which is expensive and cannot fundamentally solve *** paper provides new insights into coping with the ***,this study summarizes the false-positive problem in the mimic transformation,analyzes its possible harm and the root ***,three properties about the mimic scope are *** on the three properties and security quantification technology,the best mimic component set theory is put forward to solve the false-positive *** are two algorithms,the supplemental method and the subtraction *** best mimic component set obtained by these two algorithms can fundamentally solve the mimic system’s false-positive problem but reduce the cost of mimic *** make up for the lack of previous researches.

关键词： mimic scope the best mimic component set false-positive rate mimic transformation

来源：评论

学校读者我要写书评

暂无评论

Synthetic Voltage and Capacity Profile regeneration Through Constant Voltage Sampling Process from Second Life Batteries Perspective 10

Synthetic Voltage and Capacity Profile Regeneration Through ...

引用

10th International Conference on Electrical Energy systems, ICEES 2024

作者： Vignesh, S. Selvaraj, Jeyraj A-L Che, Hang Seng Soon, Tey Kok Universiti Malaya HiCoE UMPEDAC Level 4 WISMA R&D Institute for Advanced Studies Kuala Lumpur59990 Malaysia Universiti Malaya HiCoE UMPEDAC Level 4 WISMA R&D Kuala Lumpur59990 Malaysia Universiti Malaya Department of Computer System & Technology Kuala Lumpur50603 Malaysia

ISBN: (纸本)9798350353778

Predicting State of Health (SoH) of Second Life Batteries (SLB) is a challenging aspect when the repurposer is often surrounded by uncertainty in historical data from Electric Vehicle (EV). The EV Original Equipment Manufacturer (OEM) hardly reveal any battery related data to public domain, it also reciprocates to the repurposer. However, SoH estimation demands minimal historic data to compare past and present state of the batteries. Thus, when repurposer examines SoH of SLB, it is mandated to adhere on either of reference Performance Test (rPT) or the one which is recommended in the cell datasheet by the cell manufacturer to compare the SLB performance with the past data. While executing this process, the repurposer often conducts multiple charge and discharge cycles to comment on SoH of the SLB cells. Hence, in order to reduce the processing time in conducting cycling experiments, and to predict the SoH of SLB, the authors proposed a novel approach on regenerating voltage and capacity profile through Constant Voltage Sampling (CVS) process considering SLB perspective. The regenerated curve is deemed as synthetic curve wherein the forward and backward prediction of SoH is carried out. raw test data is accessed from University of Michigan repository for serving as reference to the authors. The results of voltage and capacity profile are validated successfully for SoH ranges from 100% to 80%. With the help of CVS technique, discharge voltage and capacity profiles for SoH ranges from 80% to 50% were reconstructed successfully to define the health indicators governing SoH and remaining useful Life (ruL). © 2024 IEEE.

关键词： State of charge

来源：评论

学校读者我要写书评

暂无评论

Parasitic Inductance of Surge Protective devices for a Wide Frequency range 37

Parasitic Inductance of Surge Protective Devices for a Wide ...

引用

37th International Conference on Lightning Protection, ICLP 2024

作者： Hadjicostas, Alexandros Chrysochos, Andreas Tsovilis, Thomas High Voltage Laboratory School of Electrical and Computer Engineering Aristotle University of Thessaloniki Thessaloniki54124 Greece R&D Department Hellenic Cables Athens15125 Greece

ISBN: (纸本)9783800764174

This work investigates the efficacy of different experimental and simulation methodologies in determining the parasitic inductance of surge protective devices for a broad frequency spectrum. This is important since the inductive behavior of surge protective devices affects their maximum residual voltage, thus also the protection offered in the case of fast-front transients exhibiting high current derivatives. The parasitic inductance of a commercially available dIN rail surge protective device is determined by utilizing a dummy surge protective device with the voltage limiting component (varistor) replaced by a copper block. Parasitic inductance estimations are made by employing: i) dielectric spectroscopy measurements, ii) impulse current tests, and iii) finite element analysis simulations;results are discussed in terms of the accuracy of the proposed methods and associated challenges in determining the parasitic inductance. © 2024 IEEE, © VdE VErLAG GMBH ∙ Berlin ∙ Offenbach.

关键词： Surge protection

来源：评论

学校读者我要写书评

暂无评论

residual Signature Analysis for Wind Turbine Failure Prognosis

Residual Signature Analysis for Wind Turbine Failure Prognos...

引用

2023 IEEE PES Innovative Smart Grid Technologies Europe, ISGT EUrOPE 2023

作者： Paul, Topon reddy, Vidhisha Ayyagari, Sai Prem Kumar Nishino, Kaneharu System AI Laboratory Corporate R&D Center Toshiba Corporation Kanagawa212-8582 Japan Pvt. Ltd KA Bangalore560034 India

ISBN: (纸本)9798350396782

In operation and maintenance of wind turbines, predictive maintenance is widely used to reduce the downtime and low operating efficiency due to failure of a wind turbine, in which the condition of a wind turbine is continuously monitored and measured by attaching a number of sensors, sign of near-future failure or degradation is detected early before the actual failure occurs, and faulty parts are replaced or maintenance is performed. In this context, automatic prognosis of failure by analyzing the time-series data of the attached sensors is very important. recently, various machine learning techniques are proposed where sign of failure is predicted based on the residual errors between the actual condition and predicted condition of a wind turbine with a prediction model built by using the time-series data. However, when the residual errors fluctuate due to external factors, this conventional approach may generate many false alarms and/or overlook potential points of failure prognosis. In this paper, we propose residual Signature Analysis (reSA) that generates anomaly scores by applying signal processing and clustering techniques on the segmented residual errors, which are generated by applying a prediction method on the time-series data. To show the effectiveness of the proposed method, we apply reSA to three wind turbines of two publicly available datasets and show the experimental results. Our experimental results suggest that our proposed method can detect sign of failure much earlier than actual time of failure with a very low rate ( © 2023 IEEE.

关键词： failure prognosis machine learning neural network operation and maintenance predictive maintenance signal processing Wind turbine

来源：评论

学校读者我要写书评

暂无评论

Fused Tapered Fan-in/Fan-out device of 6-Mode 7-Core Fiber Based On OM3 Multimode Fiber

Fused Tapered Fan-in/Fan-out Device of 6-Mode 7-Core Fiber B...

引用

2023 Asia Communications and Photonics Conference/2023 International Photonics and Optoelectronics Meetings, ACP/POEM 2023

作者： Yang, Chen du, Haoze Shao, Yuanhui Zhang, Fengming Liu, Zhuyixiao Zhang, Senyu Zhao, Zhiyong Tang, Ming R&d Center & Optics Valley Laboratory State Key Laboratory of Optical Fiber and Cable Manufacture Technology Wuhan China National Engineering Laboratory for Next Generation Internet Access System Wuhan China

ISBN: (纸本)9798350312614

Fused tapered Fan-in/Fan-out devices of few-mode multi-core fiber have received attentions in recent ten years. Most devices are based on special designed fibers which are called 'bridge fiber'. A low insertion loss and low-cost Fan-in/Fan-out device using OM3 fiber as the bridge fiber is fabricated and demonstrated in this paper. The insertion losses of LPOl, LPlla, LPllb, and LP2la are less than 1dB, and the insertion loss of LP2lb and LP02 are 2.02dB and 2.2dB. © 2023 IEEE.

关键词： Fan-in/Fan-out device few-mode multi-core fiber multimode fiber Spatial division multiplexing

来源：评论

学校读者我要写书评

暂无评论

Harnessing Machine Learning based sequence prediction for Industry 4.0 applications 15

Harnessing Machine Learning based sequence prediction for In...

引用

15th International Conference on Computing Communication and Networking Technologies, ICCCNT 2024

作者： Kanagachidambaresan, G.r. Arun, C.A. Subashini, M. Maheswaran, S. VelTech Rangarajan Dr.Sagunthala R&d institute of Science and Technology Dept of Cse Chennai India VelTech Rangarajan Dr.Sagunthala R&d institute of Science and Technology Dept of Ece Chennai India VelTec Rangarajan Dr.Sagunthala R&d institute of Science and Technology Jrf IoT Experts System Laboratory Chennai India Kongu Engineering College Department of Electronics and Communication Engineering Perundurai India

ISBN: (纸本)9798350370249

This paper focuses on the utilization of gyro sensor data, particularly from the MPU6050 sensor, for sequence prediction and positional analysis for many Industry 4.0 application and asset monitoring. Leveraging machine learning algorithms such as random Forest, decision Trees etc., a robust classification model is developed to predict sequences of positions with notable accuracy. data collection involves capturing gyro sensor readings across distinct positions, laying the foundation for subsequent analysis. Integration with the Thingsboard platform facilitates data processing and analysis, showcasing the effectiveness of this integrated approach. Additionally, a user-friendly interface is designed for streamlined data input and visualization of prediction outcomes, enhancing accessibility and usability. The model demonstrates high accuracy in predicting sequences of positions using random Forest and decision Trees algorithms. The random Forest and decision tree approach provides higher accuracy when compared with the other algorithms. A IoT based hardware is designed in Kicad and tested to validate the ML based prediction for shop floor applications. © 2024 IEEE.

关键词： data Visualization Machine Learning MPU6050 Sensor Positional Analysis Thingsboard Platform User Interface

来源：评论

学校读者我要写书评

暂无评论

Contrastive Graph representations for Logical Formulas Embedding (Extended Abstract) 40

Contrastive Graph Representations for Logical Formulas Embed...

引用

40th IEEE International Conference on data Engineering, ICdE 2024

作者： Lin, Qika Liu, Jun Zhang, Lingling Pan, Yudai Hu, Xin Xu, Fangzhi Zeng, Hongwei School of Computer Science and Technology Xi'An Jiaotong University China Xi'An Jiaotong University Shaanxi Province Key Laboratory of Satellite and Terrestrial Network Tech. R&D China

ISBN: (纸本)9798350317152

Embedding symbolic logical formulas into a low-dimensional continuous space provides an effective way for the Neural-Symbolic system. However, current studies are all constrained by the syntactic structure modeling and fail to preserve intrinsic semantics. To this end, we propose a novel model of Contrastive Graph representations (ConGr) for logical formulas embedding. Firstly, it introduces a densely connected graph convolutional network (GCN) with an attention mechanism to process syntax parsing graphs of formulas. Secondly, the contrastive instances for each anchor formula are generated by the transformation under the guidance of logical properties. Two types of contrast, global-local and global-global, are carried out to refine formula embeddings with semantic information. Extensive experiments demonstrate that ConGr obtains superior performance against state-of-the-art baselines. © 2024 IEEE.

关键词： Embeddings

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：