检索结果-内蒙古大学图书馆

Neuroadaptive Containment Control for Nonlinear Multiagent Systems With Input Saturation: An Event-Triggered Communication Approach

引用

IEEE Transactions on Systems, Man, and Cybernetics: Systems 2025年第5期55卷 3163-3173页

作者： Wang, Xin Zhang, Sen Li, Huaqing Zhang, Wei Li, Hongyi Huang, Tingwen Southwest University College of Electronic and Information Engineering Chongqing400715 China Shenzhen University of Advanced Technology Faculty of Computer Science and Control Engineering Shenzhen518055 China

In this article, a neuroadaptive event-triggered containment control strategy combined with the dynamic surface control (DSC) approach is proposed for nonlinear multiagent systems (MASs) with input saturation. Based on the event-triggered communication mechanisms, the updates of neural network weight and controllers are implemented solely under triggering conditions of violation, which markedly reduces unnecessary communication resources and minimizes inefficient control costs compared with the traditional control method. Radial basis function neural networks (RBF NNs) are employed to handle the nonlinear uncertainties of MASs. Simultaneously, an adaptive compensatory mechanism is incorporated within the backstepping process to address the nonlinear effect posed by input saturation. Additionally, we demonstrate the system stability by extending the Lyapunov theorem to jump and continuous scenarios while excluding Zeno behavior, realizing that all followers can enter the convex hull constructed by leaders. Finally, the effectiveness of the proposed methodology is verified through application simulations. © 2013 IEEE.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

Deep Subspace K-Means Clustering for Syringe Defect Detection Without Defective Samples

Deep Subspace K-Means Clustering for Syringe Defect Detectio...

引用

2023 International Symposium on Design Studies and Intelligence engineering, DSIE 2023

作者： Yan, Fei Wang, Xiaodong Hong, Longfu Xie, Zhiyao College of Computer and Information Engineering Xiamen University of Technology Xiamen China

ISBN: (纸本)9781643684888

Modern clustering techniques are becoming more popular, but traditional clustering methods often heavily rely on artificial feature extraction, costing too much human labor resources. This task is especially important when it comes to finding surface defects on medical syringes. To solve this problem, our study comes up with a new way to detect syringe defects without defective samples using deep fuzzy subspace clustering in a smooth way. In particular, our method combines conventional fuzzy c-means and deep autoencoder into a joint framework. During the clustering process, the autoencoder part automatically extract features from dataset itself and generates the representative feature embeddings. After the feature extraction, the fuzzy clustering is performed to find the optimal data structure. This "feature learning, fuzzy clustering" process is iteratively performed until the objective function converges. This new method improves the accuracy and reliability of clustering by a lot, which makes it a great choice for finding scale defects on medical syringes. The efficacy of our method is evaluated on four publicly accessible datasets and one medical syringe dataset. The results demonstrate its potential to revolutionize the field of surface defect detection in this critical domain. © 2024 The Authors.

关键词： Extraction

来源：评论

学校读者我要写书评

暂无评论

A Bilateral Assessment of Human Activities Using PSO-Based Feature Optimization and Non-linear Multi-task Least Squares Twin Support Vector Machine

引用

SN computer Science 2024年第3期5卷 315页

作者： Thakur, Ujwala Vidyarthi, Ankit Prajapati, Amarjeet Department of Computer Science Engineering and Information Technology Jaypee Institute of Information Technology Noida Sector-62 Uttar Pradesh Noida 201309 India

Human activity recognition (HAR) is an essential part of many applications, including smart surroundings, sports analysis, and healthcare. Accurately categorizing intricate actions from sensor data is still difficult, though. This method infers an individual’s actions using a variety of sensor data, including magnetometers, gyroscopes, and accelerometers. This work suggests a unique method for classifying human activities by combining a non-linear multi-task least squares twin support vector machine (NMtLSSVM) with particle swarm optimization (PSO) for feature optimization. Utilizing the advantages of both approaches, the suggested strategy achieves excellent resilience and accuracy in activity identification. The suggested method reduces dimensionality and computing costs while maintaining pertinent information using PSO to extract the most important features from the raw sensor data. On the other hand, NMtLSSVM is used to construct a multi-task learning framework that learns several related tasks at the same time and shares information among them. As a result, generalization and resilience can be enhanced beyond single-task models. Two datasets that are available to the public, WISDM, and UCI-HAR, were used to assess the suggested method. The testing findings show that the suggested approach works noticeably better than the most advanced techniques already in use. Overall activities, the outcome achieves an average classification accuracy of 97.8% (UCI-HAR dataset) and 98.5% (WISDM dataset). Furthermore, the PSO-based feature optimization decreased the feature count by 60% without sacrificing efficiency. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd 2024.

关键词： Human action recognition Intelligent computing Machine learning classification Performance comparison Sensory data analysis

来源：评论

学校读者我要写书评

暂无评论

Multi Pedestrian Tracking Algorithm Based on Improved YOLOv5 and DeepSORT 9

Multi Pedestrian Tracking Algorithm Based on Improved YOLOv5...

引用

9th International Conference on Intelligent Informatics and BioMedical Sciences, ICIIBMS 2024

作者： Wan, Hong Zhang, Yu Yang, Delong Shanghai Institute of Technology School of Computer and Information Engineering Shanghai China

ISBN: (纸本)9798350363043

A pedestrian detection and tracking algorithm based on improved YOLOv5s and DeepSORT is proposed to address the issues of pedestrian identity jumps and tracking failures caused by mutual occlusion, similar appearance features, target deformation, and motion blur on traffic roads. In the detection stage, the fusion of attention module ECA and YOLOv5 neck network enhances the extraction of pedestrian features. Replacing the CIoU bounding box loss function with the EIoU bounding box loss function accelerates bounding box regression while improving accurate localization. During the tracking phase, DeepSORT is improved by utilizing an extended Kalman filter to predict pedestrian positions in nonlinear environments. The Hungarian algorithm is used to match and detect trajectories, optimizing the problem of frequent identity switching in complex environments. Finally, the improved YOLOv5 and DeepSORT algorithms were connected to detect and track the MOT-16 dataset. The experimental results show that the improved YOLOv5 algorithm has an average accuracy increase of 3.5% compared to the original algorithm. When combined with DeepSORT tracking, the average tracking accuracy is 60.5%, which is 3.1% higher than the original *** number of pedestrian identity switches decreased by 46 times, which is 5.2% lower than the original algorithm. © 2024 IEEE.

关键词： Extended Kalman filters

来源：评论

学校读者我要写书评

暂无评论

Optimized digital watermarking: Harnessing the synergies of Schur matrix factorization, DCT, and DWT for superior image ownership proofing

引用

Multimedia Tools and Applications 2024年第18期84卷 1-36页

作者： Al-Aiash, Issa Alquran, Rabee AlJamal, Mahmoud Alsarhan, Ayoub Aljaidi, Mohammad Al-Fraihat, Dimah Department of Information Technology Faculty of Prince Al-Hussein Bin Abdallah II for Information Technology The Hashemite University Zarqa13133 Jordan Department of Computer Science Faculty of Information Technology Zarqa University Zarqa13110 Jordan Department of Software Engineering Faculty of Information Technology Isra University Amman11622 Jordan

The contemporary context of abundant digital dissemination inherently gives rise to the need for media protection and the clear identification of ownership rights. This paper responded to this nagging issue of unauthorized media capture by developing a new digital watermarking approach uniting the Discrete Wavelet Transform DWT, Discrete Cosine Transform DCT, and Schur matrix factorization. The proposed watermark, which was a positive vector uniformly distributed, presented a high degree of robustness to several adversarial operating conditions, including histogram equalization, salt and pepper noise, ripple impact, smoothing attack, and cropping. The obtained results reveal the effectiveness of our approach. The correlation coefficient between watermarked and original images remains exceptionally high at 0.999998, ensuring minimal visual distortion. Mean Squared Error (MSE) graphs for both high-frequency (HH) and low-frequency (LL) domains demonstrate negligible changes after attacks, confirming the watermark's robustness. Notably, the watermark extraction process remains successful even after histogram equalization, as indicated by significantly lower MSE compared to random vectors. Moreover, the average processing times for different attacks are promising;Histogram Equalization and Salt and Pepper attacks each took 0.3 s, Ripple Attack required 0.28 s, Smoothing Attack completed in 0.29 s, and the Cropping Attack took 0.6 s. The robustness of our approach appears against the salt and pepper attack, which historically presented a vulnerability when employing Schur matrix factorization in image watermarking, indicating the novelty of and this contribution to improving media ownership verification. By combining disparate transformation methodologies, this technique offers a promising avenue for reinforcing the integrity and authenticity of digital media, thereby strengthening the foundations of intellectual property rights in the digital domain. © The Author(s), und

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Employing Machine Learning Algorithms to Detect Phishing URL Websites 5

Employing Machine Learning Algorithms to Detect Phishing URL...

引用

5th International Conference on IoT Based Control Networks and Intelligent Systems, ICICNIS 2024

作者： Charishma, T.N.S. Koushik, A Sai Reddy, G Sai Ashmit Himabindu, M. Institute of Aeronautical Engineering Computer Science and Information Technology Hyderabad India

ISBN: (纸本)9798331518097

Phishing attacks have risen like wildfire in the era of new digital technology in the wake of exploiting holes in the electronic communication process while cheating a user to seek private information. This work provides machine learning-based resistance against phishing attacks by differentiating phishing websites using recognition of patterns and analysis on data feature. By training several machine learning models, including Random Forest, XGBoost, and Support Vector Machine (SVM), on a dataset that included characteristics frequently found on phishing websites, we developed a system that could distinguish accurately between genuine websites and phishing attempts. This project brought in a nice innovation through the incorporation of these predictive models into a Flask-developed web application, thus allowing for an interactive, real-time phishing detection. The project shall ensure protection of financial security and personal information by giving users an easy interface to input website links so they can get instant feedback about whether the link is likely to be legitimate. This hybrid technique improves accessibility to non-technical users, brings better detection accuracy, and makes phishing detection more proactive, approachable, and successful in reality. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

Evaluating Public Opinion Through Twitter Sentiment Analysis 2

Evaluating Public Opinion Through Twitter Sentiment Analysis

引用

2nd International Conference on Networking, Embedded and Wireless Systems, ICNEWS 2024

作者： Jadhav, Narendra More, Priyanka Dixit, Akhilesh Sharma, Akshay Vishwakarma Institute of Information Technology Department of Computer Engineering Pune India

ISBN: (纸本)9798350367461

In the digital age, social media platforms like Twitter have become a central forum for public discourse, reflecting diverse opinions on various topics. This paper explores the application of sentiment analysis to Twitter data to extract and quantify these opinions. This work processes and analyzes a large dataset of tweets using machine-learning techniques to identify underlying attitudes and divide them into positive, negative, and neutral categories. The methodology comprises preprocessing such as text normalization and noise removal, sentiment categorization using a model based on Natural Language Processing (NLP) techniques, and data gathering via the Twitter API. The results reveal patterns and trends in public sentiment on key issues, providing insights that can benefit businesses, policymakers, and researchers. This paper discusses the potential real-world applications of this technology in marketing, political science, and public relations. © 2024 IEEE.

关键词： Tweets

来源：评论

学校读者我要写书评

暂无评论

Quantified emotion analysis based on design principles of color feature recognition in pictures

引用

Multimedia Tools and Applications 2024年第19期83卷 57243-57267页

作者： Chen, Chin-Ling Huang, Qing-Yang Zhou, Ming Huang, Der-Chen Liu, Ling-Chun Deng, Yong-Yuan School of Information Engineering Changchun Sci-Tech University Jilin Province Changchun130600 China Department of Computer Science and Information Engineering Chaoyang University of Technology Taichung City413310 Taiwan Xiamen Academy of Arts and Design Fuzhou University Fujian Xiamen361024 China School of Computer and Information Engineering Xiamen University of Technology Fujian Xiamen361024 China Department of Computer Science and Engineering National Chung-Hsing University Taichung402 Taiwan

When people observe pictures, different pictures will generate different emotions, and the painters often convey emotional energy to the audience through the media. Through the effect of this emotional transfer, people get emotional satisfaction psychologically, and the author can better express the emotion he needs to express by using the principle. For those who do not know art appreciation, it is difficult to accurately describe the feelings brought by pictures and describe them accurately. It is also difficult to quickly and accurately find the picture of the corresponding feeling sought by oneself. This paper presents a quantitative picture emotion analysis method. Based on the analysis of the characteristics and colors of the pictures that arouse people's emotions, the classification planning is carried out in RGB space and HSV color space respectively. And the combination analysis of the color features with strong influence is carried out to select a more appropriate color space for the expression of human feelings. At the same time, the paper also discusses the effect of the black-and-white filling and the brightness saturation of the image. Finally, it makes use of the corresponding design principles to carry out a quantitative analysis of the emotion generated by it. The system performs a combined analysis of a picture from these directions and presents the results to the user. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2023.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Optimizing the Performance of the ECLAT Algorithm using K-means Clustering for Frequently Itemset Mining for Large Datasets 5

Optimizing the Performance of the ECLAT Algorithm using K-me...

引用

5th International Conference on Advanced information Technologies, ICAIT 2024

作者： Lin, Nandar Win, Thanda Yangon Technological University Computer Engineering and Information Technology Yangon Myanmar

ISBN: (纸本)9798331510046

Data mining is an approach to identify the key pattern in Frequent Itemset Mining (FIM) and to find from historical data into very useful data. ECLAT is an aspect of association rule that is applied to determine the relationship connection in item supports. By intersecting the transaction id sets according to the corresponding kth item sets based on depth-first search, ECLAT algorithm computes the support items. During searching best-selling products, ECLAT requires a lot of memory and processing time for large dataset because of the amount of transaction id sets. By merging the K-means clustering algorithm and ECLAT algorithm, the proposed method aims to determine how the K-means clustering algorithm enhances the performance of ECLAT algorithm to generate support items for large dataset. The retail data items from 100000 to 400000 UK consumers were used to retrieve the support items of the most popular selling goods. Firstly, K-means algorithm grouped the same purchased products to cluster the top selling items and then ECLAT algorithm searches the support items from clustering the famous selling products. According to the experimental results, it can be seen that the silhouette score value in this research is 0.586 to indicate a high-quality cluster, the ideal value of k is 8 clusters between 100,000 and 400,000 data items. ECLAT can search the support items based on the clustering of the top eight selling supplies by setting a minimum support value of 0.00001. The proposed method reduces memory usage and execution time to retrieve the support elements for most popular selling items than the ECLAT algorithm for massive dataset. © 2024 IEEE.

关键词： K-means clustering

来源：评论

学校读者我要写书评

暂无评论

Lightweight irregular driving behavior detection algorithm based on star operation 5

Lightweight irregular driving behavior detection algorithm b...

引用

5th International Conference on computer Vision and Data Mining, ICCVDM 2024

作者： Liang, Jiaxin Department of Computer and Information Engineering Xiamen University of Technology Xiamen China

ISBN: (纸本)9781510682863

To address the problems of high memory consumption, heavy computational burden, and difficulty deploying existing detection algorithms on edge devices, a lightweight approach for detecting abnormal driving behavior based on an updated YOLOv8 model has been presented. Firstly, the backbone network is replaced with the StarNet network, which has lower computational and memory access requirements. Second, utilizing an enhanced weighted bidirectional feature pyramid network with C2f blocks enhances the model's capability to capture multi-scale features swiftly. Finally, an asymmetric multi-level compression detection head is introduced to reduce the model size further. The experimental findings demonstrate that the suggested enhanced model achieves a 73.2% reduction in parameters, a 60.9% decrease in computational burden, and a 72.8% reduction in model size while retaining the same accuracy as the original YOLOv8 model. This significantly reduces hardware costs, demonstrating excellent lightweight effects. © 2024 SPIE.

关键词： Photons

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：