检索结果-内蒙古大学图书馆

Faster R-CNN Deep Learning Model for Pedestrian Detection from Drone Images

SN computer Science 2020年第2期1卷 1-9页

作者： Hung, Goon Li Sahimi, Mohamad Safwan Bin Samma, Hussein Almohamad, Tarik Adnan Lahasan, Badr Intelligent Biometric Group School of Electrical and Electronic Engineering Engineering Campus Universiti Sains Malaysia Penang Nibong Tebal 14300 Malaysia Department of Computer Programming Faculty of Education–Shabwa University of Aden Shabwa Yemen School of Information Science and Technology Xiamen University Malaysia Jalan Sunsuria Bandar Sunsuria Selangor Darul Ehsan Sepang 43900 Malaysia

Pedestrian detection from a drone-based images has many potential applications such as searching for missing persons, surveillance of illegal immigrants, and monitoring of critical infrastructure. However, it is considered as a very challenge computer vision problem due to the variations in camera point of view, distance from pedestrian, changes in illuminations and weather conditions, variation in the surrounding objects, as well as present of human-like objects. Recently, deep learning-based models are getting more attention, and they have proven a great success in many object detection problems such as the detection of faces, breast masses, and vehicles. As such, this work aims to develop a deep learning-based model that will be applied for the problem of pedestrian detection from a drone-based images. Particularly, faster region-based convolutional neural network (Faster R-CNN) will be used to search for the present of a pedestrian inside the captured drone-based images. To assess the performances, a total of 1500 images were collected by S30W drone and these images were captured at different places, with various views and weather conditions, and at daytime and night-time. Results show that Faster R-CNN was able to achieve a promising result with 98% precision, 99% recall, and 98% F1 measure. Further analysis has been conducted by comparing the outcomes of Faster R-CNN with YOLO deep model on UAV123 publicly available dataset. The reported results indicated that both detection models almost reported very similar results. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Deep learning Drone images Faster R-CNN Pedestrian detection

来源：评论

学校读者我要写书评

暂无评论

Automatic Detection of Smishing Attacks by Machine Learning Methods 1

Automatic Detection of Smishing Attacks by Machine Learning ...

引用

1st International Informatics and Software engineering Conference, IISEC 2019

作者： Balim, Caner Gunal, Efnan Sora Afyon Kocatepe University Dept. of Computer Programming Sandikli Vocational School Afyonkarahisar Turkey Eskisehir Osmangazi University Dept. of Computer Engineering Faculty of Engineering and Architecture Eskisehir Turkey

ISBN: (纸本)9781728139920

Phishing is a type of cyber-attack aiming to steal people's confidential information using a bait. Smishing is a form of phishing that is carried out via SMS messages sent to mobile phones. Attackers intend to steal the private information of their victims through the content they send in their SMS messages. Therefore, the detection of smishing messages has a vital role in information security. In this work, a machine learning-based model is proposed to detect smishing messages. Experiments conducted on a relatively large dataset, which contains legitimate and smishing messages, have shown that the proposed model offers a promising detection performance. © 2019 IEEE.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

A Top-Down Three-Way Merge Algorithm for HTML/XML Documents 1

引用

Science and Information Conference, SAI 2020

作者： Bakaoukas, Anastasios G. Bakaoukas, Nikolaos G. Faculty of Arts Science and Technology University of Northampton Waterside Campus University Drive NorthamptonNN1 5PH United Kingdom School of Engineering and Sciences Computer Science and Programming Department Foundation College Mitropoleos Street and Mnisikleous Campus Athens105 56 Greece

ISBN: (数字)9783030522490

ISBN: (纸本)9783030522483

Collaborative work, with the need to keep HTML/XML code up-to-date, is now becoming vital particularly in the Web Development field. In order to fully support collaborative work and resolve related problems the need has arisen for an optimum solution to the automated editing of a number of parallel copies originating from a single original HTML/XML code document with the additional requirement to subsequently merge the copies into a single updated document. A number of algorithms have been used in the past for the purpose, such as: Diff3, XmlDiff, DeltaXML & 3DM, but HTML/XML code complexity related issues have now called for an algorithm that is more specifically designed for the purpose. In this paper a new algorithmic approach to merging HTML/XML code documents is presented that is based on the "Three-way Merge" approach and the "Node-per-Node" comparison between ordered trees. For the creation of the actual merging function operating at the heart of the algorithm, a particular methodology was followed in which only the two "Current Versions" are required for the generation of the updated document, and no involvement of the "Original Document" is necessary. This is an important improvement over the currently existing algorithms because eliminates the well-known "Idempotent" problem in merging two HTML/XML documents. In addition, the algorithmic approach presented here allows for the identification and treatment of all the conflicts that arise during a HTML/XML code merging in an ordered and clearly specified manner. © 2020, Springer Nature Switzerland AG.

关键词： HTML

来源：评论

学校读者我要写书评

暂无评论

Correction: Study of the hardfacing process using self-shielding flux-cored wire with an exothermic addition with a combined oxidizer of the Al-(CuO/Fe2O3) system

引用

The International Journal of Advanced Manufacturing Technology 2024年第7期134卷 3985-3985页

作者： Trembach, Bohdan Silchenko, Yury Balenko, Oleksii Hlavchev, Dmytro Kulahin, Kostiantyn Heiko, Hennadii Bellorin-Herrera, Oleksandra Khabosha, Serhii Zakovorotnyi, Oleksandr Trembach, Illia Private Joint Stock Company “Novokramatorsky Mashinostroitelny Zavod” Kiev Ukraine Department of Computer Engineering and Programming National Technical University “Kharkiv Polytechnic Institute” Kharkiv Ukraine Scientific Research Department Ivan Kozhedub Kharkiv National Air Force University Kharkiv Ukraine

来源：评论

学校读者我要写书评

暂无评论

Methods for Using Class Based N-gram Language Models in the Kaldi Toolkit 23rd

Methods for Using Class Based N-gram Language Models in the ...

引用

23rd International Conference on Speech and computer, SPECOM 2021

作者： Pakoci, Edvin Popović, Branislav AlfaNum Speech Technologies Bulevar vojvode Stepe 40 Novi Sad21000 Serbia Faculty of Technical Sciences Department of Power Electronic and Telecommunication Engineering University of Novi Sad Trg Dositeja Obradovića 6 Novi Sad21000 Serbia Department of Music Production and Sound Design Academy of Arts Alfa BK University Nemanjina 28 Belgrade11000 Serbia Computer Programming Agency Code85 Odžaci Železnička 51 Odžaci25250 Serbia

ISBN: (纸本)9783030878016

This paper explains in detail several methods for utilization of class based n-gram language models for automatic speech recognition, within the Kaldi speech recognition framework. It reexamines an existing implementation for word-level grammars, and then presents two methods of converting ARPA-format language models into a corresponding weighted finite state transducer which is then used for decoding in Kaldi in a usual way. One of the methods relies on implicit expansion of class n-grams within the language model during the conversion into the transducer, and the other one on explicit expansion of the ARPA file by changing class n-gram entries into an array of regular class instance n-grams, while the conversion procedure remains unchanged. Some limitations and optional restrictions for the procedures are discussed as well. The proposed methods are tested on a model trained on a Serbian judicial corpus which includes classes for all types of personal nouns (first names, last names, place names, street names and organization names). © 2021, Springer Nature Switzerland AG.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

A Horizontal Federated Learning Approach to IoT Malware Traffic Detection: An Empirical Evaluation with N-BaIoT Dataset

A Horizontal Federated Learning Approach to IoT Malware Traf...

引用

International Conference on Advanced Communication Technology (ICACT)

作者： Phuc Hao Do Tran Duc Le Vladimir Vishnevsky Aleksandr Berezkin Ruslan Kirichek The Bonch-Bruevich Saint-Petersburg State University of Telecommunications Saint-Petersburg Russia Danang Architecture University Da Nang Viet Nam Department of Communication Networks and Data Transmission at Bonch-Bruevich Saint-Petersburg State University of Telecommunications Russia University of Science and Technology - The University of Danang Da Nang Viet Nam Information Technology Faculty The University of Danang - University of Science and Technology Danang Vietnam V. A. Trapeznikov Institute of Control Sciences of Russian Academy of Sciences Moscow Russia Institute of Control Sciences of Russian Academy of Sciences Russia Department of Programming Engineering and Computer Science Bonch Bruevich Saint Petersburg State University of Telecommunications

The increasing prevalence of botnet attacks in IoT networks has led to the development of deep learning techniques for their detection. However, conventional centralized deep learning models pose challenges in simultaneously ensuring user data privacy and detecting botnet attacks. To address this issue, this study evaluates the efficacy of Federated Learning (FL) in detecting IoT malware traffic while preserving user privacy. The study employs N-BaIoT, a dataset of real-world IoT network traffic infected by malware, and compares the effectiveness of FL models using Convolutional Neural Network, Long Short-Term Memory, and Gated Recurrent Unit models with a centralized approach. The results indicate that FL can achieve high performance in detecting abnormal traffic in IoT networks, with the CNN model yielding the best results among the three models evaluated. The study recommends the use of FL for IoT malware traffic detection due to its ability to preserve data privacy.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Structure and Functioning of Information Systems of Background Monitoring of Landscape Elements of Gorgany Nature Reserve 10

Structure and Functioning of Information Systems of Backgrou...

引用

10th International Conference on Advanced computer Information Technologies, ACIT 2020

作者： Nykolaychuk, Yaroslav Petrashchuk, Yaroslav Slobodian, Olena Pitukh, Ihor Grynchyshyn, Taras Nykolaychuk, Lyubov Hryha, Volodymyr Ternopil National Economic University Department of Specialized Computer Systems Ternopil Ukraine Gorgany Nature Reserve Nadvirna Ukraine Nadvirna College by National Transport University Department of Programming and Information Technologies Nadvirna Ukraine Ivano-Frankivsk National Technical University of Oil and Gas Department of Social Communication and Law Ivano-Frankivsk Ukraine Vasyl Stefanyk Precarpathian National University Department of Computer Engineering and Electronics Ivano-Frankivsk Ukraine

This paper discusses the structural organization of the background monitoring system (BMS) of the transboundary UNESCO World Heritage Site 'Beech forests of the Carpathians and ancient beech forests of Germany' using actual data and information from Gorgany Nature Reserve in Ivano-Frankivsk, Ukraine. The features of the structure and functions of background monitoring of the protected natural resources, forests and wildlife of Gorgany Nature Reserve is based on modern methods of collecting and processing of digital data, implementation of the principles of Internet of Things and cloud technologies, combined with GPS-systems, sensor networks with the use of optical and fiber-optic telecommunication channels. The information technology of the background monitoring system is based on figurative cluster models of monitoring objects according to theoretical principles of statistical, correlatory, spectral, cluster, logic-statistical and entropy analysis. © 2020 IEEE.

关键词： Sensor networks

来源：评论

学校读者我要写书评

暂无评论

A non-potential Target Function for Controlling the UAVs Group Flight in Presence of Concave Obstacles

A non-potential Target Function for Controlling the UAVs Gro...

引用

IEEE International Conference on Actual Problems of Unmanned Air Vehicles Developments Proceedings (APUAVD)

作者： Olexiy M. Glazok Computer and programming engineering National Aviation University Kyiv Ukraine

ISBN: (纸本)9781728125930

A solution for the problem of controlling the unmanned quadrotor vehicles group flight in presence of obstacles of complex form is proposed. A traditional approach to construction of the target function does not solve the problem of control at presence of concave obstacles; the UAVs may get trapped in such a situation. The proposed target function contains the three terms. The first of them controls the UAVs' behavior with respect to attitude. The second is built on the base of a moving coordinate system bound to the trajectory of the group motion. The third addend is a standard sum that performs the function of repelling the UAVs one from another. Also, in the proposed approach the motion is bound to the virtual center of the group, instead of one UAV that is chosen to be a leader. This approach helps to avoid a discrepancy between desired and actual states of the group in case that the leader UAV experiences some problems with its motion.

关键词： Unmanned aerial vehicles Trajectory Acceleration Mathematical model Collision avoidance Standards Autonomous vehicles

来源：评论

学校读者我要写书评

暂无评论

Modelling the Decision Making Process on Legal Conducting the Surrogate Motherhood

Modelling the Decision Making Process on Legal Conducting th...

引用

IEEE International Conference on Dependable Systems, Services and Technologies (DESSERT)

作者： Tetiana Hovorushchenko Alla Herts Yelyzaveta Hnatchuk Computer Engineering & System Programming Department Khmelnytskyi National University Khmelnytskyi Ukraine Civil Law and Legal Proceedings Department Ivan Franko National University of Lviv Lviv Ukraine

ISBN: (数字)9781728199573

ISBN: (纸本)9781728199580

Currently, there is a need in Ukraine for the design and development of multidisciplinary decision support systems (DSSs) for the medical and other related spheres - for example, the medical law sector, in particular, the DSS for the surrogacy's legal regulation field. This DSS will be able to protect doctors from potentially wrong decisions by taking into account all available information when making the decisions. For design such DSS, it's necessary to conduct the modelling of the decision-making process on surrogacy's legal conducting, that is the goal of this research. In this paper, there are the theoretical foundations of using the ontologies for the decision-making process on surrogacy's legal conducting, a model of base ontology with the necessary requirements and recommendations for conducting the surrogate motherhood, and ontology-based model of the decision-making process on surrogacy's legal conducting. The conducted modelling is a theoretical basis for the development of methods and design of DSS in the field of legal regulation of surrogate motherhood.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Modern Information Technologies in Scientific Research and Educational Activities

arXiv

引用

arXiv 2024年

作者： Malakhov, Kyrylo Kaverinskiy, Vladislav Ivanova, Liliia Romanyuk, Oleksandr Romaniuk, Oksana Voinova, Svitlana Kotlyk, Sergii Sokolova, Oksana V. M. Glushkov Institute of Cybernetics The National Academy of Sciences of Ukraine Kyiv Ukraine Department of Wear-Resistant and Corrosion-Resistant Powder Construction Materials Frantsevic Institute for Problems in Material Science The National Academy of Sciences of Ukraine Kyiv Ukraine Separated Structural Subdivision «Odesa Technical Professional College of Odesa National University of Technology» Odessa Ukraine Vinnytsia National Technical university Vinnytsia Ukraine Vinnytsia National Technical university Vinnytsia Ukraine Odesa National University of Technology Odessa Ukraine The Institute of Computer Engineering Automation Robotics and Programming Odesa National University of Technology Odessa Ukraine Odesa National University of Technology Odessa Ukraine

This article presents a comprehensive methodology for enhancing the capabilities of conversational AI systems, focusing on ChatGPT, through the integration of ontology-driven structured prompts and meta-learning techniques. The proposed approach, exemplified by the RDWE OntoChatGPT system, demonstrates significant potential for improving productivity and expanding knowledge bases in specific subject domains and languages. By formalizing methods for creating and configuring structured prompts and context ontologies, the system enables ChatGPT to provide specific information using selected contexts, thereby enhancing its performance in handling data across diverse linguistic and domain contexts. Through empirical testing, the study evaluates the system’s effectiveness in providing accurate responses while also highlighting areas for further refinement. The methodology’s adaptability to other conversational AI systems based on large language models underscores its potential for broader application in natural language processing and dialogue systems. Future enhancements aim to refine structured prompts and incorporate additional input data to further improve system performance and accuracy. Overall, this research contributes to advancing the capabilities of conversational AI systems, paving the way for more sophisticated and effective human-computer interactions in various domains. © 2024, CC BY-SA.

关键词： Ontology

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：