检索结果-内蒙古大学图书馆

Cat Swarm Algorithm Generated Based on Genetic Programming Framework Applied in Digital Watermarking

computers, Materials & Continua 2025年第5期83卷 3135-3163页

作者： Shu-Chuan Chu Libin Fu Jeng-Shyang Pan Xingsi Xue Min Liu School of Artificial Intelligence Nanjing University of Information Science and TechnologyNanjing210000China College of Computer Science and Engineering Shandong University of Science and TechnologyQingdao266000China Department of Information Management Chaoyang University of TechnologyTaichung40601TaiwanChina School of Information Science and Engineering Fujian University of TechnologyFuzhou350000China Key Laboratory of Hydrometeorological Disaster Mechanism and Warning of Ministry of Water Resource Nanjing University of Information Science and TechnologyNanjing210000China

Evolutionary algorithms have been extensively utilized in practical ***,manually designed population updating formulas are inherently prone to the subjective influence of the *** programming(GP),characterized by its tree-based solution structure,is a widely adopted technique for optimizing the structure of mathematical models tailored to real-world *** paper introduces a GP-based framework(GPEAs)for the autonomous generation of update formulas,aiming to reduce human *** modifications to tree-based GP have been instigated,encompassing adjustments to its initialization process and fundamental update operations such as crossover and mutation within the *** designing suitable function sets and terminal sets tailored to the selected evolutionary algorithm,and ultimately derive an improved update *** Cat Swarm Optimization Algorithm(CSO)is chosen as a case study,and the GP-EAs is employed to regenerate the speed update formulas of the *** validate the feasibility of the GP-EAs,the comprehensive performance of the enhanced algorithm(GP-CSO)was evaluated on the CEC2017 benchmark ***,GP-CSO is applied to deduce suitable embedding factors,thereby improving the robustness of the digital watermarking *** experimental results indicate that the update formulas generated through training with GP-EAs possess excellent performance scalability and practical application proficiency.

关键词： Cat swarm algorithm genetic programming digital watermarking update mode mode generation framework

来源：评论

学校读者我要写书评

暂无评论

A survey of video violence detection

引用

Cyber-Physical Systems 2023年第1期9卷 1-24页

作者： Yao, Huiling Hu, Xing School of Optical-Electrical Information and Computer Engineering University of Shanghai for Science and Technology Shanghai China

As one of the important applications of intelligent video surveillance, violent behaviour detection (VioBD) plays a crucial role in public security and safety. As a particular type of behaviour recognition, VioBD aims to identify whether the behaviours that occurred in the scene is aggressive, such as fighting and assault. To comprehensively analyse the current state and predict the future trend of VioBD research, we survey the existing approaches of VioBD in this work. First, we briefly introduce the basic principle and the challenges of VioBD;Then, we category the existing approaches according to their framework, including the traditional framework, end-to-end deep learning framework, and hybrid deep learning framework. Finally, we introduce the public datasets for evaluating the performance of VioBD approaches and compare their performances on these datasets. Besides, we also summarise the open problems in VioBD and predict its future trends. © 2021 Informa UK Limited, trading as Taylor & Francis Group.

关键词： Surveys

来源：评论

学校读者我要写书评

暂无评论

Botnet Detection Through Flow-Based Deep Feature Extraction and Ensemble Classification

引用

Journal of The Institution of Engineers (India): Series B 2025年 1-14页

作者： Geetha, K. Brahmananda, S.H. Department of Computer Science and Engineering GITAM School of Technology Karnataka Bengaluru India Ashtaksha Labs Pvt Ltd Karnataka Bengaluru India

As Internet of Things (IoT) devices are networked and thus susceptible to many forms of attacks, cyber security risk is the primary concern in the IoT field. To tackle this issue, this study employs machine learning and deep learning techniques to identify botnet attacks. The proposed method uses a random forest classifier or other machine learning techniques to organize the data after extracting detailed characteristics using Bidirectional Long Short-Term Memory (Bi-LSTM). This methodology yields superior outcomes for challenges involving binary and multiclass classification. The IoT-23 and N-BaIoT databases were used. The IoT-23 collection consists of three normal samples and twenty malware samples from different IoT devices, whereas the N-BaIoT dataset includes scenarios like Mirai and Bashlite attacks. In the context of IoT risks, the proposed technique successfully differentiates between benign and malicious traffic data. In terms of multiclass classification, the accuracy rates for N-BaIoT and the IoT-23 dataset are 100% and 99.9%, respectively, while the acquired accuracy rates for binary classification are 99.9% for N-BaIoT and 97% for the IoT-23 dataset. © The Institution of Engineers (India) 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Q-learning based strategy analysis of cyber-physical systems considering unequal cost

引用

Intelligent and Converged Networks 2023年第2期4卷 116-126页

作者： Xin Chen Jixiang Cheng Luanjuan Jiang Qianmu Li Ting Wang Dafang Li School of Management Science and Engineering Nanjing University of Finance and EconomicsNanjing 210023China School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210014China

This paper proposes a cyber security strategy for cyber-physical systems(CPS)based on Q-learning under unequal cost to obtain a more efficient and low-cost cyber security defense strategy with misclassification *** system loss caused by strategy selection errors in the cyber security of CPS is often considered ***,sometimes the cost associated with different errors in strategy selection may not always be the same due to the severity of the consequences of ***,unequal costs referring to the fact that different strategy selection errors may result in different levels of system losses can significantly affect the overall performance of the strategy selection *** introducing a weight parameter that adjusts the unequal cost associated with different types of misclassification errors,a modified Q-learning algorithm is proposed to develop a defense strategy that minimizes system loss in CPS with misclassification interference,and the objective of the algorithm is shifted towards minimizing the overall ***,simulations are conducted to compare the proposed approach with the standard Q-learning based cyber security strategy method,which assumes equal costs for all types of misclassification *** results demonstrate the effectiveness and feasibility of the proposed research.

关键词： cyber security Q-learning policy selection unequal cost misclassification interference

来源：评论

学校读者我要写书评

暂无评论

Classifying aquatic organism via evolving residual convolutional neural networks by optimized random vector functional link

引用

Multimedia Tools and Applications 2025年第1期84卷 221-243页

作者： Zhou, Zhiyu Wen, Jingqi Liu, Mingxuan Yang, Junyi School of Computer Science and Technology Zhejiang Sci-Tech University Hangzhou310018 China School of Mechanical Engineering Hangzhou Dianzi University Hangzhou310018 China

Aquatic organisms serve as crucial indicators of ecosystem health and water quality conditions. Accurate classification and monitoring of aquatic organisms facilitate the timely detection of ecological environmental changes, which is of significant importance for the conservation of biodiversity and environmental protection. Pointing at the issues of fuzzy images and difficult classification caused by the poor lighting conditions of aquatic organisms and many impurities around, we put forward a novel classification model of aquatic organisms with convolution neural network, feedforward network and optimization algorithm. The model first abstracts the traits of aquatic organisms, and then classifies the abstracted features. Through migration learning of ResNet101 neural network on Fish4Knowledge dataset and ISLVRC2012 and some data sets related to aquatic organisms, a network that will be used to abstract features is obtained. The model put forward in the paper firstly uses ResNet101 to abstract the aquatic organism image’s characteristics, and then uses the Golden Eagle optimization algorithm (GEO) optimized by chaos mapping optimization and the RVFL neural network optimized by L2,1 regularization to classify aquatic organisms. The algorithm is compared with other RVFL algorithms optimized by Ant Lion Optimizer (ALO), Grey Wolf Optimization (GWO), Particle Swarm Optimization (PSO), Harris Hawks Optimization (HHO) and Dragonfly Algorithm (DA). The findings indicate that the methodology introduced in this paper exhibits an improvement over the original ResNet101, with increases of 1.58%, 2.57%, 0.22% and 0.93% in particularity rate, accuracy rate, recall rate and F1 score on ILSVRC2012, respectively. The efficiency and accuracy of this method significantly contribute to the advancement of aquatic organism monitoring and environmental protection efforts. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Secure location-aware geocast routing in Internet of Vehicles

引用

Multimedia Tools and Applications 2025年第18期84卷 19669-19692页

作者： Kumar, Sunil Dohare, Upasana Kumar, Sushil Department of Computer Science and Information Technology Central University of Haryana Mahendragarh India School of Computing Science & Engineering Galgotias University Greater Noida India School of Computer and Systems Sciences Jawaharlal Nehru University New Delhi India

Efficient message dissemination in Vehicular Ad Hoc Networks (VANETs) relies on robust connectivity between neighboring vehicular nodes, yet it is often compromised by malicious intruders. While recent literature proposes various approaches to enhance geocasting performance, security considerations for connectivity are often overlooked. To address this gap, we propose the Secure Location-Aware Cache Agent based Geocast Routing (SL-CAGR) protocol, emphasizing an Intrusion Prevention Technique (IPT) for message integrity preservation and a Connectivity Reliant Packet Forwarding (CRPF) algorithm for improved VANET connectivity. Our protocol introduces a network model for cache agent-based geocast routing on the Internet of Vehicle (IoV) environment, implements IPT using Diffie-Hellman digital signatures, and presents a secure geocast routing strategy focusing on re-caching and cache agent selection. Empirical simulations demonstrate SL-CAGR's superiority in packet delivery ratio, packet loss, duplication, hop count, end-to-end delay, and network load compared to these protocols. Simulation results demonstrate that SL-CAGR outperforms state-of-the-art techniques like CAG, CTRC, and D-Flooding. SL-CAGR stands out by merging intrusion prevention, secure geocast routing, and cache agent selection, ensuring robust security and efficiency in VANETs. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Vehicular ad hoc networks

来源：评论

学校读者我要写书评

暂无评论

Human novel pose generation with pose optimization and importance sampling based on NeRF 16

Human novel pose generation with pose optimization and impor...

引用

2024 16th International Conference on Graphics and Image Processing, ICGIP 2024

作者： Li, Ke Zhang, Binbin Su, Zhiyong Li, Weiqing School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing China School of Automation Nanjing University of Science and Technology Nanjing China

ISBN: (数字)9781510688780

ISBN: (纸本)9781510688773

The Neural Radiance Field (NeRF) method has emerged as a groundbreaking technique for human reconstruction, enabling the generation of high-quality, photorealistic rendering of reconstructed objects. Despite its promising results, several challenges persist, particularly when it comes to generalizing to unknown poses during the generation of virtual human animations. This paper proposes a method for generating and optimizing novel human poses based on NeRF. Given that existing pose-driven NeRF methods heavily rely on the accuracy of SMPL body parameters, we implemente human pose optimization based on motion capture data as prior knowledge, resulting in more accurate SMPL parameters, which contributes to a more precise human canonical model. Additionally, we propose a pose-driven deformation field based on linear blend skinning, combining blend weight fields with the 3D human skeleton to achieve precise mapping from observed poses to the canonical pose. By estimating the camera ray transmittance distribution during volume rendering, we propose importance sampling method avoiding sampling points in free space, which further improves rendering quality. Extensive experiments demonstrate that our method yields significant improvements in rendering quality for both training poses and novel poses, as evidenced by results on the ZJU-MoCap dataset and our custom dataset. This work not only advances the capabilities of NeRF in human reconstruction but also opens new avenues for realistic virtual human animation. © 2025 SPIE.

关键词： Motion capture

来源：评论

学校读者我要写书评

暂无评论

Efficient Ship:A Hybrid Deep Learning Framework for Ship Detection in the River

引用

computer Modeling in engineering & sciences 2024年第1期138卷 301-320页

作者： Huafeng Chen Junxing Xue Hanyun Wen Yurong Hu Yudong Zhang School of Computer Engineering Jingchu University of TechnologyJingmen448000China School of Computer Science Yangtze UniversityJingzhou434023China School of Computing and Mathematic Sciences University of LeicesterLeicesterLE17RHUK

Optical image-based ship detection can ensure the safety of ships and promote the orderly management of ships in offshore *** deep learning researches on optical image-based ship detection mainly focus on improving one-stage detectors for real-time ship detection but sacrifices the accuracy of *** solve this problem,we present a hybrid ship detection framework which is named EfficientShip in this *** core parts of the EfficientShip are DLA-backboned object location(DBOL)and CascadeRCNN-guided object classification(CROC).The DBOL is responsible for finding potential ship objects,and the CROC is used to categorize the potential ship *** also design a pixel-spatial-level data augmentation(PSDA)to reduce the risk of detection model *** compare the proposed EfficientShip with state-of-the-art(SOTA)literature on a ship detection dataset called *** show our ship detection framework achieves a result of 99.63%(mAP)at 45 fps,which is much better than 8 SOTA approaches on detection accuracy and can also meet the requirements of real-time application scenarios.

关键词： Ship detection deep learning data augmentation object location object classification

来源：评论

学校读者我要写书评

暂无评论

Feature selection for determining input parameters in antenna modeling

引用

Journal of Systems engineering and Electronics 2025年第1期36卷 15-23页

作者： LIU Zhixian SHAO Wei CHENG Xi OU Haiyan DING Xiao School of Physics University of Electronic Science and Technology of ChinaChengdu 611731China School of Computer and Information Engineering Xinjiang Agricultural UniversityUrumqi 830052China

In this paper,a feature selection method for determining input parameters in antenna modeling is *** antenna modeling,the input feature of artificial neural network(ANN)is geometric *** selection criteria contain correlation and sensitivity between the geometric parameter and the electromagnetic(EM)*** information coefficient(MIC),an exploratory data mining tool,is introduced to evaluate both linear and nonlinear *** EM response range is utilized to evaluate the *** wide response range corresponding to varying values of a parameter implies the parameter is highly sensitive and the narrow response range suggests the parameter is *** the parameter which is highly correlative and sensitive is selected as the input of ANN,and the sampling space of the model is highly *** modeling of a wideband and circularly polarized antenna is studied as an example to verify the effectiveness of the proposed *** number of input parameters decreases from8 to *** testing errors of|S_(11)|and axis ratio are reduced by8.74%and 8.95%,respectively,compared with the ANN with no feature selection.

关键词： antenna modeling artificial neural network(ANN) feature selection maximal information coefficient(MIC)

来源：评论

学校读者我要写书评

暂无评论

Automated Resume Parsing and Ranking using Natural Language Processing 3

Automated Resume Parsing and Ranking using Natural Language ...

引用

3rd International Conference on Artificial Intelligence For Internet of Things, AIIoT 2024

作者： Thangaramya, K. Logeswari, G. Gajendran, Sudhakaran Deepika Roselind, J. Ahirwar, Neha School of Computer Science and Engineering Vellore Institute of Technology Vellore India School of Computer Science and Engineering Vellore Institute of Technology Chennai Campus India School of Electronics Engineering Vellore Institute of Technology Chennai Campus India

ISBN: (纸本)9798350372120

Advancements in networking and communication have revolutionized the recruitment process, leading to the development of modern resume parsing and ranking systems. With the proliferation of internet-based recruiting, numerous resumes are now stored in recruitment systems. However, traditional methods, such as manual processing and the utilization of unique resume templates, have limitations when dealing with unstructured documents like resumes. Resume parsing is a crucial technique that involves extracting essential details pertaining to applicants to be shortlisted for interview. This paper introduces the 'Resume Parser and Ranker,' an application designed to efficiently and automatically rank resumes, saving considerable time and manpower during the recruitment process. The application is powered by Natural Language Processing (NLP) approach, incorporates heuristic calculations to evaluate the final score of each candidate. The system utilizes Deep Learning (DL) for Named Entity Recognition (NER), achieving an impressive 93% accuracy in information extraction. Notably, this accuracy is close to the level achieved by humans, which typically does not exceed 96%. The high accuracy of the resume parser ensures reliable results, enabling companies to identify the most suitable candidates who can be called for further interview rounds. © 2024 IEEE.

关键词： Deep learning

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：