检索结果-内蒙古大学图书馆

The importance of feature selection in online signature verification

Neural Computing and Applications 2024年 1-13页

作者： Szucs, Cintia Lia Kovari, Bence Department of Automation and Applied Informatics Budapest University of Technology and Economics Budapest Hungary

Handwritten signatures are one of the most commonly used biometrics. Because signatures are widely accepted, their verification is a fundamental problem. Online signature verification applies different electronic devices to capture the signatures, which allow the use of dynamic information, such as the pressure or inclination angle of the pen. Therefore, it is much more challenging to forge online signatures. Further derived features can be defined and calculated based on the captured features. The selection of applied features is a crucial step of signature verification. This work aims to analyse the importance of feature selection by comparing the usability of common derived function features using dynamic time warping (DTW) based and Siamese bidirectional long short-term memory (BiLSTM)-based solutions. This paper comprehensively studies the most common function features and their literature background, presents several feature sets including pairs and triplets, compares their effectiveness using different classification approaches, and presents a DTW-based classifier and four Siamese BiLSTM models using two different loss functions. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

A Self-adapting Pixel Antenna - Substrate Lens System for Infrared Frequencies

Periodica Polytechnica Electrical Engineering and Computer S...

引用

Periodica Polytechnica Electrical Engineering and Computer Science 2025年第1期69卷 14-25页

作者： Shubbar, Mustafa Rakos, Balázs Department of Automation and Applied Informatics Faculty of Electrical Engineering and Informatics Budapest University of Technology and Economics Műegyetem rkp. 3. BudapestH-1111 Hungary

In this work, we propose a concentrator lens - antenna arrangement, potentially suitable for infrared sensing and energy harvesting rectenna systems. The structure consists of a pixel antenna, self-adapting for the direction of the incident infrared radiation, and a concentrator lens, both optimized for the mid-infrared spectrum. The silicon substrate lens is situated above the pixel antenna, enabling the concentration of the incident light on the antenna;silicon was chosen due to its transparency in the IR spectrum. We examine how various parameters of the lens, in conjunction with the different states of the pixel antenna, affects the performance of the system. The simulations show that the gain of the arrangement increases considerably in correlation with the radius of the lens. The results suggest that the energy conversion efficiency of infrared rectenna systems can be enhanced by several orders of magnitude with the utilization of the proposed arrangement. © 2025 Budapest University of Technology and Economics. All rights reserved.

关键词： Rectennas

来源：评论

学校读者我要写书评

暂无评论

Optimization-based Model Predictive Tube Control for Autonomous Ground Vehicles with Minimal Tuning Parameters

引用

无人系统（英文） 2023年第1期11卷 93-108页

作者： Adorjan Kovacs Istvan Vajk Department of Automation and Applied Informatics Faculty of Electrical Engineering and InformaticsBudapest University of Technology and EconomicsMegyetem rkp.3H-1111 BudapestHungary

The trajectory planning and tracking problem are critical points of intelligent vehicles concerning their safety and *** these parts are separated,interactions should be made between them,especially when sudden changes and disturbances *** paper presents a method that integrates the two parts using a cascade ***,the proposed method deals with the interaction between the lateral and the longitudinal trajectory based on dynamical *** whole problem is handled using the model predictive method based on online *** system receives the path borders as input and generates the control requests for the actuators on its *** configuration space of the system can be maximized to gain stability by handling the lateral-longitudinal parts and the trajectory planning and tracking in one complex *** main advantage of the proposed approach is that the optimization problem in the predictive method is formulated so that the path and dynamics are considered equality and inequality constraints,and the cost function includes only a physical phenomenon to be minimized without tuning *** evaluation of the proposed algorithm is presented in this paper based on simulation and real-time measurements results.

关键词： Path planning path following autonomous car convex optimization model predictive control internal model control

来源：评论

学校读者我要写书评

暂无评论

Real-Time Evaluation of Weighting Factorless Predictive Control of LCL Filter Equipped Grid-Side Converters using Sorting Networks

Real-Time Evaluation of Weighting Factorless Predictive Cont...

引用

2024 International Exhibition and Conference for Power Electronics, Intelligent Motion, Renewable Energy and Energy Management, PCIM Europe 2024

作者： Bándy, Kristóf Stumpf, Péter Süto, Zoltán Department of Automation and Applied Informatics Faculty of Electrical Engineering and Informatics Budapest University of Technology and Economics Hungary

ISBN: (纸本)9783800762620

A novel multi-cost function model-based finite-set predictive control method is presented for grid-side converters equipped with LCL filters. This approach aims to overcome the tedious tuning of weighting factors. In the method, the cost functions are established based on various state variables and their references. Then the control options are ranked based on their performance in each of the cost functions. This ranking score determines the selection of the optimal control action. The control method is validated in a low-cost custom Hardware-in-the-Loop (HIL) environment using model-based rapid-prototyping tools of Matlab/Simulink. Simulation and measurement results show the viability of the method. © VDE VERLAG GMBH · Berlin · Offenbach.

关键词： Hardware-in-the-loop simulation

来源：评论

学校读者我要写书评

暂无评论

Machine Learning and Dynamic Time Warping for Online Signature Verification 32

Machine Learning and Dynamic Time Warping for Online Signatu...

引用

32nd International Conference on Software, Telecommunications and Computer Networks, SoftCOM 2024

作者： Saleem, Mohammad Budapest University of Technology and Economics Department of Automation and Applied Informatics Budapest Hungary

ISBN: (纸本)9789532901382

Signature verification is a very common task in the field of document analysis used for the identification and validation of individuals. Many approaches can be used for signature classification. Using a machine-learning-based system is one of the most common approaches. This paper uses machine learning to classify signatures as genuine or forged. It also used a dynamic time-warping approach in the preprocessing step to produce data for training purposes. This approach differs from using it in the classification phase, which is a common practice in signature verification. The work targets online signatures, which are obtained using digital devices. Different databases were used to evaluate the accuracy of the verification systems, and four different algorithms were used for classification. The results are very promising. The minimum error rate achieved was 2.09% for the MCYT-100 dataset, 3.34% for the SVC2004 dataset, 1. 6 8 % for the SigComp'11-Chinese dataset, and 3. 6 8 % for the SigComp'11- Dutch dataset. © 2024 University of Split, FESB.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Enhanced Indoor Mapping via Integration of Lidar and HoloLens 2 Data Using Iterative Closest Point Algorithm 22

Enhanced Indoor Mapping via Integration of Lidar and HoloLen...

引用

22nd IEEE International Symposium on Intelligent Systems and informatics, SISY 2024

作者： Szalka, Panka Hideg, Attila Budapest University of Technology and Economics Faculty of Electrical Engineering and Informatics Department of Automation and Applied Informatics Budapest Hungary

ISBN: (纸本)9798350385601

This study presents an innovative method to improve the accuracy of indoor mapping by deploying and integrating a sensor that transmits two-dimensional data with a three-dimensional mesh created by another device. The Iterative Closest Point algorithm was used to coordinate the different data sets and transform their different coordinate systems into a common indoor map. Prior to this, rigorous image cleaning and matching processes are conducted to ensure the quality of the data used. The effectiveness of this methodology has been validated through multiple tests using a Lidar scanner and a HoloLens 2 Augmented Reality device. This method not only offers potential for enhancing indoor navigation and safety protocols like emergency evacuation but also presents opportunities for advancing asset tracking and other applications. © 2024 IEEE.

关键词： Indoor positioning systems

来源：评论

学校读者我要写书评

暂无评论

Speech Emotion Recognition Using openSMILE and GPT 3.5 Transformer

Speech Emotion Recognition Using openSMILE and GPT 3.5 Trans...

引用

34th Medical informatics Europe Conference, MIE 2024

作者： Turcian, Darius Stoicu-Tivadar, Vasile Politehnica University Timișoara Department of Automation and Applied Informatics Timișoara Romania

ISBN: (纸本)9781643685335

In recent years, artificial intelligence, and machine learning (ML) models have advanced significantly, offering transformative solutions across diverse sectors. Emotion recognition in speech has particularly benefited from ML techniques, revolutionizing its accuracy and applicability. This article proposes a method for emotion detection in Romanian speech analysis by combining two distinct approaches: semantic analysis using GPT Transformer and acoustic analysis using openSMILE. The results showed an accuracy of 74% and a precision of almost 82%. Several system limitations were observed due to the limited and low-quality dataset. However, it also opened a new horizon in our research by analyzing emotions to identify mental health disorders. © 2024 The Authors.

关键词： GPT 3.5 Machine Learning openSMILE Speech Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Low Latency Video Streaming System for VR Teleoperation over 5G Networks 3

Low Latency Video Streaming System for VR Teleoperation over...

引用

3rd IEEE International Conference on Metrology for eXtended Reality, Artificial Intelligence and Neural Engineering, MetroXRAINE 2024

作者： Blázovics, László Kovács, Viktor Gótzy, Márton Budapest University of Technology and Economics Faculty of Electrical Engineering and Informatics Department of Automation and Applied Informatics BudapestH-1111 Hungary

ISBN: (纸本)9798350378009

With the introduction of 5G networks, high-speed low-latency network connections became available for mobile and embedded devices without the limited range of WiFi networks. Although the source and the destination can now be hundreds of kilometres away, the communication between them can be flawless. Our goal is to have a low cost, low power, lightweight device that is able to stream video for UAV (unmanned aerial vehicles), mobile robot teleoperation over 5G networks to AR/VR (augmented/virtual reality) applications, to large distances without the need for line of sight. In this paper we compare the latency of existing low cost video streamer solutions (UAV camera streaming, single board computer) and present a custom hardware-software solution optimized for low latency, while conducting measurements over different network types (wired, campus 5G, public 5G). We have shown that the custom device was able to achieve with significantly lower latency (50-80 ms direct over direct connection, about 100-150 ms over public 5G). © 2024 IEEE.

关键词： 5G mobile communication systems

来源：评论

学校读者我要写书评

暂无评论

Using Weka API for creating a custom classification application 17

Using Weka API for creating a custom classification applicat...

引用

17th IEEE International Symposium on applied Computational Intelligence and informatics, SACI 2023

作者： Robu, Raul Arseni-Ailoi, Paul Ungureanu-Anghel, Dan Politehnica University of Timisoara Faculty of Automation and Computers Department of Automation and Applied Informatics Timisoara Romania

ISBN: (纸本)9798350321104

In real world applications, classification models can be built by repeatedly going through the steps of data preprocessing, building classification models with different dedicated algorithms, testing the resulted models, until building a good enough classification model is achieved. The model built in this way can be used in order to perform predictions on new data, with a degree of trust resulted by testing the model. All the operations described above can be done with Weka, which is a very powerful machine learning tool. The authors consider that after going through the previous steps, for many real world applications, it would be beneficial to go through a supplementary step, of developing a custom application for data classification, specific to the realized study. The proposed application will use the Weka API, which will allow it to rebuild the selected classification model, save and load it in/from a binary file and then make predictions with the help of a user interface perfectly adapted to the data set used. The proposed application makes it easier to make predictions for a final beneficiary. The paper presents how such a custom classification application can be developed, using the Weka API and Java programming language. In order to validate the proposed solution, the authors have built a custom application for the classification of data of patients who may have heart diseases, starting from the Cleveland heart disease data set, obtained from the UCI Machine Learning Repository. The proposed application allows saving and loading the classification model through serialization and deserialization, so that building it in order to make predictions is necessary only once. © 2023 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Cloud-Based Digital Twin for Robot Health Monitoring and Integration in Cyber-Physical Production Systems 13th

Cloud-Based Digital Twin for Robot Health Monitoring and Int...

引用

Joint International Conference of the 13th IFToMM International Symposium on Science of Mechanisms and Machines, SYROM 2022 and 25th International Conference on Robotics, ROBOTICS 2022

作者： Borangiu, Th. Răileanu, S. Anton, F. Lențoiu, I. Negoiță, R. Department of Automation and Applied Informatics University Politehnica of Bucharest Bucharest Romania

ISBN: (纸本)9783031256547

The paper describes an aggregate cloud-based digital twin (DT) that models an industrial robot by virtualizing its physical structure, motion and action control, operating modes, parameters and interactions with shop floor devices. The DT functionalities are: tuning robot motion parameters, shop floor layout validation for collision avoidance, robot health monitoring and anomaly detection, concurrent job allocation to multiple robots, by implementing in cloud IAAS three DT-specific software configurations: simulation with software in the loop;robust deployment;embedded simulation faster than real-time. Experimental results are reported from data collecting ACE software for Omron-Adept robots. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Cyber Physical System

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：