检索结果-内蒙古大学图书馆

2024 International Conference on Optimization Computing and Wireless Communication, ICOCWC 2024

作者： Yadav, Dhananjay Kumar Kumari, Neeraj Harron, Syed Maharishi School of Engineering & Technology Maharishi University of Information Technology Uttar Pradesh India Teerthanker Mahaveer University Department of Computing Sciences and Information Technology Uttar Pradesh Moradabad India Department of Computer Science and Information Technology Bangalore India

ISBN: (纸本)9798350383348

Convolutional neural networks (CNNs) have emerged as a powerful tool for object detection and recognition. Recent advances in CNNs have improved their performance on object detection by incorporating innovative convolutional layers and architectures. These advances include the inception architecture, region proposal networks (RPNs), and fully convolutional networks (FCNs). Additionally, these architectures have enabled object detection and recognition with significant improvements in accuracy and speed. Furthermore, recent research has focused on applying deep transfer learning techniques to CNNs for object detection and recognition, which have shown promising results in terms of precision and accuracy. Overall, these ongoing advancements have further improved the state of the art in object detection and recognition tasks. © 2024 IEEE.

关键词： Additionally Architectures Convolutional Improvements Layers Promising Recognition

来源：评论

学校读者我要写书评

暂无评论

Strategies for Helping Anchor-Based Trackers Learn re-ID Features for Smart City Surveillance

Strategies for Helping Anchor-Based Trackers Learn re-ID Fea...

引用

2024 IEEE International Conference on Consumer Electronics, ICCE 2024

作者： Chen, Xiu-Zhi Li, Mu-Chuan Chen, Yen-Lin National Taipei University of Technology Dept. of Computer Science & Information Engineering Taipei Taiwan

ISBN: (纸本)9798350324136

Re-identification has become a crucial issue in computer vision today as it allows for tracking objects in both continuous and discontinuous scenarios. Despite achieving perfect detection results, anchor-based trackers encountered difficulties in effectively learning re-identification features, due to various issues. This research proposes strategies aimed at improving the capability of anchor-based trackers to learn high-quality re-identification (re-ID) features. The model developed through our strategies can extract more distinct features and achieve almost 0.57 Multiple Object Tracking Accuracy (MOTA) on MOT20, even under a limited training dataset. This result indicates that our proposed strategies hold potential for improving the performance of anchor-based trackers. © 2024 IEEE.

关键词： Anchor-Based Tracker Re-Identification Smart City Surveillance

来源：评论

学校读者我要写书评

暂无评论

An Efficient Convolutional Multi-Scale Vision Transformer for Image Classification 2

An Efficient Convolutional Multi-Scale Vision Transformer fo...

引用

2nd International Conference on Image Processing, computer Vision and Machine Learning, ICICML 2023

作者： Zhang, Ji Chen, Zhihao Ge, Yiyuan Yu, Mingxin Beijing Information Science and Technology University School of Instrument Science and Opto-Electronics Engineering Beijing China Beijing Information Science and Technology University Computer School Beijing China Jilin University Department of Software Engineering Changchun China

ISBN: (纸本)9798350331417

This paper introduces an innovative and efficient multi-scale Vision Transformer (ViT) for the task of image classification. The proposed model leverages the inherent power of transformer architecture and combines it with the concept of multi-scale processing generally used in convolutional neural networks (CNNs). The work aims to address the limitations of conventional ViTs which typically operate at a single scale, hence overlooking the hierarchical structure in visual data. The multi-scale ViT enhances classification performance by processing image features at different scales, effectively capturing both low-level and high-level semantic information. Extensive experimental results demonstrate the superior performance of the proposed model over standard ViTs and other state-of-the-art image classification methods, signifying the effectiveness of the multi-scale approach. This research opens new avenues for incorporating scale-variance in transformer-based models for improved performance in vision tasks. © 2023 IEEE.

关键词： Classification Convolutional neural networks Multi-scale Vision Transformer

来源：评论

学校读者我要写书评

暂无评论

Maintaining Mobile Communication in Distress and Emergency Situations 1

引用

32nd International Conference on Organization and technology of Maintenance, OTO 2023

作者： Barić, Tomislav Glavaš, Hrvoje Faculty of Electrical Engineering Computer Science and Information Technology Osijek University of Osijek Osijek Croatia

ISBN: (数字)9783031514944

ISBN: (纸本)9783031514937

This paper provides an overview of the technical aspects of the application of amateur radio stations in maintaining mobile communication in distress and emergency situations. Ultra-high frequencies (UHF) were selected for radio communication testing. Testing was performed at a frequency of 446 MHz, which is covered by the PMR446 (Private Mobile Radio) standard. Handheld radios were used to test communication in urban conditions. Given the high attenuation of electromagnetic waves in urban areas due to concrete and steel reinforced concrete buildings, special attention is paid to testing communication in urban areas. The quality and range of communication using non-directional and directional antennas as well as with simplex repeaters are presented in the paper. The obtained measurement results of quality and range of communication in urban conditions are analyzed and interpreted. General conclusions are given on the capabilities of amateur radio handheld stations and additional equipment such as antennas and simplex repeaters for maintaining communication in distress and emergency situations in urban conditions. © 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

关键词： Electromagnetic waves

来源：评论

学校读者我要写书评

暂无评论

TransProtFam: A Transformer Derived Deep Learning Model for Hierarchical Protein Family Annotation

TransProtFam: A Transformer Derived Deep Learning Model for ...

引用

2024 IEEE International Conference on computer Vision and Machine Intelligence, CVMI 2024

作者： John, Chinju Sahoo, Jayakrushna Indian Institute of Information Technology Kottayam Department of Computer Science and Engineering Kerala Pala India

ISBN: (纸本)9798350376876

Transformers are knowledge reservoirs trained on vast and heterogeneous data sets with the potential for cross-functional tasks. These large corpus-trained models can be leveraged for cumbersome tasks like Protein family annotations which are pivotal for discerning the features of a protein sequence and its functional roles. Here we have harnessed the knowledge derived from such protein sequence-based transformers to fine-tune them for protein family *** performance of such models was compared on benchmark datasets like GPCR and PoG to annotate protein sequences at hierarchical levels. Parameter Efficient Fine Tuning was used to optimise the number of trainable parameters and to perform less computationally expensive training. Our proposed model -TransProtFam could come up with accuracy scores of 99.43%,90.85% and 84.49% at family, sub-family and sub-subfamily levels, which exceeds the state-of-the-art neural network models. Our findings and the finetuning process we employed may provide insights into utilising large models trained on protein sequence corpora for specific tasks in Bioinformatics and related domains. © 2024 IEEE.

关键词： Distribution transformers

来源：评论

学校读者我要写书评

暂无评论

Do Current Video LLMs Have Strong OCR Abilities? A Preliminary Study 31

Do Current Video LLMs Have Strong OCR Abilities? A Prelimina...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Fei, Yulin Gao, Yuhui Xian, Xingyuan Zhang, Xiaojin Wu, Tao Chen, Wei School of Software Engineering Huazhong University of Science and Technology China Information Hub Hongkong University of Science and Technology Guangzhou China

ISBN: (纸本)9798891761964

With the rise of multi-modal large language models, accurately extracting and understanding textual information from video content-referred to as video-based optical character recognition (Video OCR)-has become a crucial capability. This paper introduces a novel benchmark designed to evaluate the video OCR performance of multi-modal models in videos. Comprising 1,028 videos and 2,961 question-answer pairs, this benchmark proposes several key challenges through 6 distinct sub-tasks: (1) Recognition of text content itself and its basic visual attributes, (2) Semantic and Spatial Comprehension of OCR objects in videos (3) Dynamic Motion detection and Temporal Localization. We developed this benchmark using a semi-automated approach that integrates the OCR ability of image LLMs with manual refinement, balancing efficiency, cost, and data quality. Our resource aims to help advance research in video LLMs and underscores the need for improving OCR ability for video LLMs. The benchmark will be released on https://***/YuHuiGao/***. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Methods for Detecting Android Malware: Employing Mobile Devices to Improve Procedures for Inquiry-Based Learning

International Journal of Interactive Mobile Technologies

引用

International Journal of Interactive Mobile Technologies 2025年第6期19卷 103-114页

作者： Yelisetti, Sandeep Suresh, P. Jha, Suresh Kumar Arora, Neha Anand, Ketan Sailaja, Vedala Naga Information Technology Velagapudi Ramakrishna Siddhartha Engineering College Deemed to be University Andhra Pradesh Vijayawada India Department of Computer Science and Engineering Saveetha School of Engineering SIMATS Tamil Nadu Thandalam Chennai India Manipal University Rajasthan Jaipur India School of Information Science Presidency University Karnataka Bengaluru India Telangana Hyderabad India Dept of MBA KLEF Deemed to be University Green Fields Andhra Pradesh Vaddeswaram India

The operating system (OS) of a computer controls both its hardware and software. It handles necessary functions including input and output processing, file and memory management, and peripheral device management, including disc drives and printers. Programs created for particular purposes are referred to as application software. These programs, which are frequently open source and freely accessible, contribute to the growing number of downloads. This paper discusses the basics of Android malware, its evolution, and malware analysis tools and techniques. Providing the research gaps and giving a review of the literature on Android malware detection using machine learning and deep learning are its main objectives. It offers the knowledge gathered from the literature as well as potential avenues for future research, which may aid in the development of reliable and precise methods for classifying Android malware. This paper conducts a systematic and comprehensive assessment of the methods and tools utilized for the analysis, classification, and detection of malicious Android apps. Several research gaps are indicated based on the thorough literature evaluation. Additionally, the report offers insights on future research paths that may aid academics in developing novel and reliable methods for identifying and categorizing Android malware. © 2025 by the authors of this article.

关键词： Android malware

来源：评论

学校读者我要写书评

暂无评论

On the Smoothed Complexity of Combinatorial Local Search 51

On the Smoothed Complexity of Combinatorial Local Search

引用

51st International Colloquium on Automata, Languages, and Programming, ICALP 2024

作者： Giannakopoulos, Yiannis Grosz, Alexander Melissourgos, Themistoklis School of Computing Science University of Glasgow United Kingdom School of Computation Information and Technology Technical University of Munich Germany School of Computer Science and Electronic Engineering University of Essex United Kingdom

ISBN: (纸本)9783959773225

We propose a unifying framework for smoothed analysis of combinatorial local optimization problems, and show how a diverse selection of problems within the complexity class PLS can be cast within this model. This abstraction allows us to identify key structural properties, and corresponding parameters, that determine the smoothed running time of local search dynamics. We formalize this via a black-box tool that provides concrete bounds on the expected maximum number of steps needed until local search reaches an exact local optimum. This bound is particularly strong, in the sense that it holds for any starting feasible solution, any choice of pivoting rule, and does not rely on the choice of specific noise distributions that are applied on the input, but it is parameterized by just a global upper bound on the probability density. The power of this tool can be demonstrated by instantiating it for various PLS-hard problems of interest to derive efficient smoothed running times (as a function of and the input size). Most notably, we focus on the important local optimization problem of finding pure Nash equilibria in Congestion Games, that has not been studied before from a smoothed analysis perspective. Specifically, we propose novel smoothed analysis models for general and Network Congestion Games, under various representations, including explicit, step-function, and polynomial resource latencies. We study PLS-hard instances of these problems and show that their standard local search algorithms run in polynomial smoothed time. Further applications of our framework to a wide range of additional combinatorial problems can be found in the full version of our paper. © Yiannis Giannakopoulos, Alexander Grosz, and Themistoklis Melissourgos.

关键词： Probability distributions

来源：评论

学校读者我要写书评

暂无评论

Exploring the Development of Flexible Network Subsystems in Communications Software Design 15

Exploring the Development of Flexible Network Subsystems in ...

引用

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

作者： Mamatha, G.N. Pandeya, Megha Saraswat, Bhawna School of Engineering and Technology Department of Electronics and Communication Engineering Karnataka Bangalore India Maharishi University of Information Technology Maharishi School of Engineering & Technology Uttar Pradesh India Sanskriti University Department of Computer Science Engineering Uttar Pradesh Mathura India

ISBN: (纸本)9798350370249

The development of cutting-edge telecommunications systems more and more call for the implementation of community subsystems which might be bendy and can unexpectedly reply to modifications in user visitors. Such flexibility requires the layout of a structure able to scaling to distinctive gadget demands while nevertheless being capable of maintain excessive overall performance, reliability, and availability. This paper gives a systematic exploration of the strategies, gear, and ideas important in designing a bendy community subsystem for telecommunications software. We speak the necessities and challenges of developing network structures, and have a look at the feasibility of employing current trendy protocols which includes SIP, SDP, and RTP as a part of a usual structure. Furthermore, we speak the advantages and downsides of using software program-defined networking (SDN) as a part of a flexible community subsystem, along with issues for deploying it. Updates to present community control structures to assist dynamic reconfiguration are explored, in addition to an assessment of in-house and 1/3-celebration answers for integration into a bendy community subsystem. Sooner or later, key layout issues for deploying and managing a flexible community subsystem are discussed and summarized. © 2024 IEEE.

关键词： Downsides Implementation Necessities Networking Nevertheless Telecommunications

来源：评论

学校读者我要写书评

暂无评论

Biomedical signal processing for health monitoring applications: a review

引用

International Journal of Applied Systemic Studies 2023年第1期10卷 44-69页

作者： Sonawani, Shilpa Patil, Kailas Natarajan, Prabhu School of Computer Engineering and Technology Dr. Vishwanath Karad MIT World Peace University India Department of Computer Engineering Vishwakarma University India Department of Computer Science DigiPen Institute of Technology Singapore

Biomedical health monitoring systems are evolving rapidly and using non-invasive and cost effective sensors. These systems can monitor physiological parameters of the body to monitor health conditions and provide feedback. These new generation systems can use advance technologies to build intelligent systems for health monitoring and timely diseases detection and diagnosis. Biomedical signal processing and analysis of the patterns in the signals plays important role in building efficient systems, life sciences and research and is rapidly expanding in this domain. This review paper is significant as no existing review paper is giving complete information on biomedical signal processing phases as a whole. It makes four contributions;first, it gives activity flow for developing biomedical signal processing systems. Second, reviews various recent applications researched and kinds of low cost, non-invasive sensors used for biomedical health monitoring. Third, categorisation and enlisting of signal processing techniques like signal sampling, segmentation, filtering, feature extraction, dimensionality reduction and machine learning techniques in healthcare domain. Fourth, it reviews challenges in using signal processing techniques and gives future perspectives which can be helpful for the researchers to use advance techniques to build intelligent systems for reliable decision making. Copyright © 2023 Inderscience Enterprises Ltd.

关键词： Cost effectiveness

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：