检索结果-内蒙古大学图书馆

3rd International Conference on Artificial Intelligence and Signal Processing, AISP 2023

作者： Lasrado, Rohan Nolan Vijayasherly, V. VIT University School of Computer Science and Engineering Vellore India VIT University Department of Software Systems Vellore India

ISBN: (纸本)9798350320749

Neural generative modelling of sketches has been an active research direction. SketchRNN set a milestone with their sequence-to-sequence variational autoencoder architecture being able to generate hand drawn sketches in various classes by modelling them as sequences of displacements between consecutive stroke points. The diversity and variety in the set of handwritten Chinese characters makes them a good candidate for such generative modelling enabling their unconditional generation. However, modelling them as sequences of points causes coarse looking strokes and much longer sequences, consequently requiring use of polygonal approximation algorithms to cut down on points. Instead, we propose and investigate the modelling of Chinese characters as sequences of Bézier curves using the SketchRNN architecture with a few modifications, to allow the model to directly generate smooth curves. This way the encoded representation is smaller while more of the stroke's characteristics are retained and the generated characters are truly scalable. We also suggest the appropriate preprocessing strategy for the KanjiVG dataset to make it suitable for this purpose. Qualitative evaluation of the results suggests the model demonstrates generation of characters with mostly well-structured and ordered strokes. This was substantiated by quantitative evaluation based on the FrFréchetchet Inception Distance score. © 2023 IEEE.

关键词： Learning systems

来源：评论

学校读者我要写书评

暂无评论

Freeze-Phish: An ANN Based Phishing Detection System

引用

International Journal of Network Security 2023年第5期25卷 893-898页

作者： Yang, Cheng-Ying Shih, Chun-Yi Yang, Chou-Chen Hwang, Min-Shiang Department of Computer Science University of Taipei Taipei Taiwan Department of Management Information Systems National Chung Hsing University Taichung Taiwan Department of Computer Science and Information Engineering Asia University Taichung41354 Taiwan Fintech and Blockchain Research Center Asia University Taichung41354 Taiwan

Phishing (also known as online phishing) is an online criminal. Attackers use a fake webpage that imitates trusted websites to steal sensitive personal information such as passwords and credit card details. In this paper, we propose an application named Freeze-Phish, which uses Python to build a web crawler to collect information such as hyperlinks from the website. In addition, we build a brand word and suspicious word database by editing distance algorithms like Levenshtein distance and Hamming distance to compare the difference between the words in the website URL and the suspicious one. Then, we use a neural network to train our model and export our code as executable code(.exe) so that our users can use our code more easily to detect suspicious websites. Compared to other methods, our model’s accuracy is about 97% true positive rate, and the average execution time is 21.3 seconds. © (2023). All Rights Reserved.

关键词： Hamming distance

来源：评论

学校读者我要写书评

暂无评论

Tolerant Tracking Control Protocol for PMSM Based on Policy Iteration Algorithm and Fault Compensation 16th

Tolerant Tracking Control Protocol for PMSM Based on Polic...

引用

16th International Conference on Social Robotics, ICSR + InnoBiz 2024

作者： Yan, Shuya Li, Xiaocong Qian, Huaming Ma, Jun Al Mamun, Abdullah College of Intelligent Systems Science and Engineering Harbin Engineering University Harbin150001 China Department of Electrical and Computer Engineering National University of Singapore Singapore117583 Singapore Guangzhou China Information Science and Technology Eastern Institute of Technology Ningbo China

ISBN: (纸本)9789819611508

In robotics applications, ensuring reliable performance in the presence of actuator faults is essential for maintaining system safety and reliability. This paper presents a tolerant tracking control method for permanent magnet synchronous motors (PMSMs) based on adaptive dynamic programming and fault compensation. The method simultaneously considers tracking accuracy and energy consumption through a policy iteration algorithm. In the optimality analysis of the algorithm, more relaxed conditions are provided to demonstrate that the performance function can converge to a near-optimal value within a finite number of iterations. In practical implementation, an actor-critic network is used to approximate the performance function and control protocol, alongside a fault detection mechanism based on an expanded time horizon, which achieves fault detection from arbitrary initial values. The effectiveness of the proposed algorithm is verified using a high-fidelity PMSM model in Simulink. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Permanent magnets

来源：评论

学校读者我要写书评

暂无评论

Latency Reduction In Industrial Iot Using Yolov8 CNN Approach For Real-Time Data Processing 15

Latency Reduction In Industrial Iot Using Yolov8 CNN Approac...

引用

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

作者： Nirmala, G. Prasanthi, Yeddula Sajitha, L.P. Ramesh Babu, P. Pratheeksha Hegde, N. Laxmipriya, S. R.M.D Engineering College Department of Computer Science and Engineering Chennai India College of Agricultural Engineering Andhra Pradesh Madakasira India R. M. K Engineering College Department of Computer Science and Business Systems Tamilnadu Kavaraipettai India Wollega University College of Engineering and Technology Department of Computer Science Oromia Region Nekemte Ethiopia Nitte Deemed to be University Department of Ise Nmamit Karnataka Udupi India Mgr Educational and Research Institute Department of Civil Engineering Maduravayol Chennai India

ISBN: (纸本)9798350370249

During the era of Industry 4.0, the incorporation of the Internet of Things (IoT) into various industrial processes has become increasingly widespread, resulting in increased levels of production and efficiency. On the other hand, while dealing with the huge amounts of data produced by Industrial Internet of Things (IIoT) devices in real-time, it is particularly difficult to minimise latency in order to guarantee rapid decision-making capabilities. There has been a significant increase in the amount of data generated in industrial settings as a result of the proliferation of IIoT devices. Low-latency data processing is something that businesses that rely on timely insights need to achieve in order to improve their operational efficiency, optimise their operations, and prevent failures. Through the application of the YoLov8 Convolutional Neural Network (CNN), this research introduces a novel approach to the processing of data in industrial settings in real time. The goal of this study is to find a solution to this urgent problem. The proposed approach involves utilising the YoLov8 CNN architecture to manage data in real time inside an Internet of Things environment. Through the use of simulated industrial scenarios, the study examines the latency reduction capabilities of YoLov8 in comparison to those of more conventional techniques. When it comes to processing data in real time in an industrial setting, the results demonstrate that the YoLov8 CNN approach is effective in significantly reducing the amount of latency that occurs. © 2024 IEEE.

关键词： Convolutional Neural Network Industrial IoT Latency reduction Real-time data YoLov8 CNN

来源：评论

学校读者我要写书评

暂无评论

Multidimensional Voronoi Constellations vs. Short Blocklength Probabilistic Shaping: A Comparison for Multilevel Coding Approach

Multidimensional Voronoi Constellations vs. Short Blocklengt...

引用

2024 Asia Communications and Photonics Conference, ACP 2024 and International Conference on Information Photonics and Optical Communications, IPOC 2024

作者： Sheng, Yajie Chen, Bin Lei, Yi Deng, Jingxin Xu, Jiwei Fu, Mengfan Zhuge, Qunbi Li, Shen Department of Computer Science and Information Engineering Hefei University of Technology Hefei China State Key Laboratory of Advanced Optical Communication Systems and Networks Department of Electronic Engineering Shanghai Jiao Tong University Shanghai China Department of Electrical and Computer Engineering Université Laval Québec Canada

Performance of concatenated multilevel coding with probabilistic shaping (PS) and Voronoi constellations (VCs) is analysed over AWGN channel. Numerical results show that VCs provide up to 1.3 dB SNR gains over PS-QAM ... 详细信息

ISBN: (纸本)9798350379266

关键词： Probability distributions

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Energy and Secure Efficient Routing for Wireless Sensor Networks

Performance Analysis of Energy and Secure Efficient Routing ...

引用

2024 International Conference on Decision Aid sciences and Applications, DASA 2024

作者： Ahmed, Amjed Abbas Hussein, Zainab Mohammed Najim, Ali Hamza Ali Alheeti, Khattab M. Mohd Satar, Nurhizam Safie Abdallah Hashim, Aisha Hassan Universiti Kebangsaan Malaysia Center for Cyber Security Faculty of Information Science and Technology Bangi Malaysia Department of Computer Techniques Engineering Baghdad Iraq Baghdad Iraq Department of Computer Technical Engineering Al-Diwaniyah Iraq College of Computer and Information Technology Computer Networking Systems Department University of Anbar Anbar Iraq Center for Cyber Security Faculty of Information Science and Technology Bangi43600 Malaysia International Islamic University Malaysia Faculty of Engineering Department of Electrical and Computer Engineering Selangor Malaysia

ISBN: (纸本)9798350369106

Wireless Sensor Networks (WSNs) are essential for various applications, but their architecture makes them vulnerable to attacks. While traditional security methods like authentication and encryption offer some protection, they often fall short against specific node attacks. This paper introduces the Secure Energy Aware Routing (SEAR) strategy, which uses trust-based assessments to exclude malicious nodes while optimizing energy consumption. SEAR factors in node trust, hop time, and residual energy to ensure reliable data transmission through trustworthy nodes. MATLAB simulations on networks of varying sizes (100-500) compare SEAR with existing methods (EAD and EASER), demonstrating SEAR's superior performance. Results show significant improvements in packet delivery ratio (13-16%), network throughput (19-24 %), energy consumption (32-36%), packet loss (16-190/0), and end-to-end latency (21-26%), proving SEAR's effectiveness in enhancing WSN security, energy efficiency, and performance. © 2024 IEEE.

关键词： MATLAB

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Movie Recommendation systems Using Filtering Techniques on IMDB and Rotten Tomatoes 15

Comparative Analysis of Movie Recommendation Systems Using F...

引用

15th International Conference on Information, Intelligence, systems and Applications, IISA 2024

作者： Majhi, Santosh Kumar Sahani, Preetipadma Pradhan, Rosy Acharya, Biswaranjan Grivokostopoulou, Foteini Kanavos, Andreas Gerogiannis, Vassilis C. Department of Computer Science and Information Technology Bilaspur India Veer Surendra Sai University of Technology Department of Computer Science and Engineering Odisha Burla India Veer Surendra Sai University of Technology Department of Electrical Engineering Odisha Burla India Marwadi University Department of Computer Engineering Ai Rajkot India Computer Technology Institute and Press 'Diophantus' Patras Greece Ionian University Department of Informatics Corfu Greece University of Thessaly Department of Digital Systems Larissa Greece

ISBN: (纸本)9798350368833

Recommendation systems are pivotal in industries like entertainment, where competitiveness is intense. This study enhances movie recommendations by utilizing rich datasets from IMDB and Rotten Tomatoes, incorporating movie ratings, user profiles, audience scores, and critics' reviews. These elements were subjected to sentiment analysis to refine the data quality. We explored three filtering techniques: content-based, which analyzes movie features;collaborative, which processes user behavior patterns;and hybrid, combining both to leverage their strengths. The effectiveness of these methods was assessed using Mean Average Precision at k (MAP-k) and Root Mean Square Error (RMSE), with our results showing notable improvements in recommendation accuracy. For instance, the hybrid method achieved a MAP ©k of 0.271 and an RMSE of 0.634, indicating enhanced precision and reliability of the recommendations. These findings suggest significant potential for improving user satisfaction and retention in digital platforms, demonstrating the benefit of integrating diverse data sources and analytical techniques in recommendation systems.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

Oct Image Blind Despeckling Based on Gradient Guided Filter with Speckle Statistical Prior 48

Oct Image Blind Despeckling Based on Gradient Guided Filter ...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Li, Sanqian Xiong, Muxing Yang, Bing Zhang, Xiaoqing Higashita, Risa Liu, Jiang Southern University of Science and Technology Research Institute of Trustworthy Autonomous Systems Department of Computer Science and Engineering Shenzhen China Tomey Corporation Japan

ISBN: (纸本)9781728163277

Optical coherence tomography (OCT) imaging technique has been widely used for ocular disease diagnosis. However, speckles occur in OCT images due to the property of coherent imaging, inevitably affecting the visual quality and clinical analysis. To alleviate this problem, we propose a novel gradient-guided speckle image filtering method (GGSF) with structure enhancement for directly removing speckles in OCT images. Specifically, the multiplicative characteristic of speckle noise is incorporated into the guided filtering processing for modeling raw OCT images. To avoid getting trapped in image distortions, we further employ gradient regularization to integrate the structure prior information into the guided speckle image filtering procedure. Additionally, we introduce the statistical property of speckle noise obeying a gamma distribution into the least square method solver for the resulting non-convex GGSF model. Experimental results on the AS-OCT dataset demonstrate the effectiveness of GGSF for OCT image despeckling compared with competitive methods. Furthermore, we validate the benefits of GGSF for subsequent clinical analysis with the CM-OCT dataset. © 2023 IEEE.

关键词： Optical tomography

来源：评论

学校读者我要写书评

暂无评论

Pediatric Bone Age Prediction Using Deep Learning 26

Pediatric Bone Age Prediction Using Deep Learning

引用

26th International Conference on computer and Information Technology, ICCIT 2023

作者： Bin Habib, Al Zadid Sultan Islam, Md. Ekramul Bin Syed, Md Asif Ahamed, Md Younus Tasnim, Tanpia West Virginia University Lane Department of Computer Science and Electrical Engineering MorgantownWV26506 United States Stamford University Bangladesh Department of Computer Science & Engineering Dhaka1217 Bangladesh West Virginia University Department of Industrial and Management Systems Engineering MorgantownWV26506 United States Green University of Bangladesh Department of Computer Science and Engineering Dhaka1461 Bangladesh

ISBN: (纸本)9798350359015

Pediatric bone age prediction is a crucial task in clinical practice that can help diagnose endocrine disorders and provide insight into a child's growth and development. However, conventional bone age prediction methods are often labor-intensive and require specialized radiological expertise. This paper presents a Deep Learning (DL)-based approach to pediatric bone age prediction using EfficientNet with Additive Attention, a state-of-the-art neural network architecture for image classification and regression tasks. The method utilizes over 12,000 X-ray images from the RSNA bone age dataset. It involves image preprocessing, transforming them into three-channel images, and training a Convolutional Neural Network (CNN) to automatically learn the features of hand bone images. This approach provides a more effective and accurate solution for predicting bone age, which is critical in diagnosing pediatric endocrine diseases. This work uses two variations of the EfficientNet model (B0 and B4), where EfficientNetB4 is also finetuned with the Additive Attention mechanism. These three models predict the age for the original age, and their comparison is shown in curves. The predicted ages depict that in most cases, EfficientNetB4 and EfficientNetB4 with Additive Attention (EN-AA) successfully predicted the bone ages more accurately regarding the original age, and their performance was better than the EfficientNetB0. Specific performance metrics are provided to underscore this improvement. Learning curves for training and validation loss confirm effective learning without overfitting or underfitting, further validating our approach's efficacy in pediatric endocrine disease diagnosis. © 2023 IEEE.

关键词： Additive Attention Deep Learning EfficientNet Image Regression Pediatrics RSNA Bone Age

来源：评论

学校读者我要写书评

暂无评论

ATFS: A deep learning framework for angle transformation and face swapping of face de-identification

引用

Multimedia Tools and Applications 2024年第12期83卷 36797-36822页

作者： Tsai, Chwei-Shyong Wu, Hsien-Chu Chen, Wei-Ting Ying, Josh Jia-Ching Department of Management Information Systems National Chung Hsing University Taichung402 Taiwan Language Center National Chin-Yi University of Technology Taichung411 Taiwan Department of Computer Science and Information Engineering National Taichung University of Science and Technology Taichung404 Taiwan

This paper proposes an effective framework for improving angle transformation and face replacement. This ATFS framework aims to solve the image distortion caused by angled face swapping in face de-identification. In the ATFS framework, TransNet and SWGAN were proposed based on generative adversarial networks. When TransNet uses neural networks to reconstruct images, it combines the predicted facial points. As a result, Transnet can locate the target’s facial features and generate multi-angle transformed images to achieve angle transformation. The SWGAN applies a complicated neural network that combines residual operations and self-attention modules to extract the input image features and replace the face region of the input image with the target image to achieve face de-identification. Both TransNet and SWGAN adopt discriminators and various loss functions to train the neural network and accurately accelerate the training process. The experimental results show that the proposed method can maintain high-quality images and avoid image distortion during face swapping for image angle transformation compared to previously proposed methods. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2023.

关键词： Generative adversarial networks

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：