检索结果-内蒙古大学图书馆

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

作者： Mandale, Prashant Modi, Shabina Jadhav, Makarand Mohan Khawate, Swati Shailesh Borate, Vishal Kisan Mali, Yogesh Kisan Department of Information Technology Pune India Karmaveer Bhaurao Patil College of Engineering Satara India Dept of Electronics & Telecommunication Engg NBN SinhgadTechnical Institutes Campus Pune India Ajeenkya D Y Patil School of Engineering Dept Electronics and Telecommunication Pune India Dr D Y Patil College of Engineering & Innovation Department of Computer Engineering Talegaon Pune India G.H Raisoni College of Engineering and Management Department of Artificial Intelligence Wagholi Pune India

ISBN: (纸本)9798350370249

The arising development of the Web of Things (IoT) offers different adaptability in information move processes, specialized strategies, and capacity. distributed storage is used by limit of the populace to use their information for simple access. Then again, business utilization of distributed storage additionally expanding. The Cloud control is produced through a brought together organization. The arising development gives an approach to outsider access in normally associated networks. digital actual frameworks(CPS) are applied in many computerized networks. The investigation of different difficulties in digital actual frameworks and their effect on information security are examined in different academic articles. The arising development of AI innovation offers solid advantages for digital actual frameworks security. The introduced concentrate on led far reaching research on difficulties in digital actual frameworks and planned an idea for future improvement. © 2024 IEEE.

关键词： Cloud security

来源：评论

学校读者我要写书评

暂无评论

Skin Cancer disease Classification Using Tasnet V2 In Image Classification 2

Skin Cancer Disease Classification Using Tasnet V2 In Image ...

引用

2nd International Conference on disruptive Technologies, ICdT 2024

作者： Banu, M.Sheerin Elanangai, V. Chakraborty, Uddyalok Sharmiladevi, d. Ramesh Babu, P. Rajeswari, P. R.M.K. Engineering College Department of Information Technology India St. Peter's Institute of Higher Education and Research Department of Eee Avadi Tamilnadu Chennai India Upl University of Sustainable Technology Department of Computer Engineering Gujarat Ankleshwar India Sona College of Technology Tamilnadu Salem India College of Engineering and Technology Wollega University Department of Computer Science Oromia Region Nekemte Ethiopia R.M.D Engineering College India

ISBN: (纸本)9798350371055

Skin cancer is very common and can be fatal. To improve patient results, it is essential to find the disease early. It can be hard to tell the difference between benign and malignant skin lesions when using standard picture classification methods. The complicated features and intricate patterns that can be seen in dermatoscopic images make it very hard to use the current methods for classifying skin cancer signs. If a diagnosis is delayed or not dealt with properly, the effects can be very bad. The study goal is to make skin cancer classification more accurate with the help of TasNet v2 in order to meet a need that appears in the current methods. This study suggests that TasNet v2, a state-of-the-art deep learning framework, be used to sort skin cancer into different groups. In order to solve the problem of the need for more reliable and advanced diagnostic tools, this would work. The suggested method uses transfer learning on a very large set of labelled dermatoscopic images to teach TasNet v2 how to find signs of skin cancer. The trained model is checked and tried on several different sets of data. Using TasNet v2 unique design, which improves feature extraction, lets the model find smaller details that could be signs of skin cancer. The fact that TasNet v2 was able to do picture classification tasks that were not in its original area of expertise shows how flexible and useful it is. These results make it more likely that a new development will happen that will help doctors make more accurate skin cancer diagnoses. © 2024 IEEE.

关键词： deep learning dermatoscopic Images Image Classification Skin Cancer TasNet v2

来源：评论

学校读者我要写书评

暂无评论

Revisiting lARS for large Batch Training Generalization of Neural Networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第5期6卷 1321-1333页

作者： do, Khoi Nguyen, Minh-duong Hoa, Nguyen Tien Tran-Thanh, long Tran, Nguyen H. Pham, Quoc-Viet Trinity College Dublin School of Computer Science and Statistics Dublin 2 D02PN40 Ireland Pusan National University Department of Information Convergence Engineering Busan46241 Korea Republic of Hanoi University of Science and Technology School of Electrical and Electronic Engineering Hanoi100000 Viet Nam University of Warwick School of Computer Science CoventryCV4 7AL United Kingdom The University of Sydney School of Computer Science DarlingtonNSW2006 Australia Kyung Hee University Department of Computer Science and Engineering Yongin446-701 Korea Republic of

This article investigates large batch training techniques using layer-wise adaptive scaling ratio (lARS) across diverse settings. In particular, we first show that a state-of-the-art technique, called lARS with the warm-up, tends to be trapped in sharp minimizers early on due to redundant ratio scaling. Additionally, a fixed steep decline in the latter phase restricts deep neural networks from effectively navigating early-phase sharp minimizers. To address these issues, we propose time varying lARS (TVlARS), a novel algorithm that replaces warm-up with a configurable sigmoid-like function for robust training in the initial phase. TVlARS promotes gradient exploration early on, surpassing sharp optimizers and gradually transitioning to lARS for robustness in later stages. Extensive experiments demonstrate that TVlARS consistently outperforms lARS and lAMB in most cases, with up to 2% improvement in classification scenarios. In all self-supervised learning cases, TVlARS achieves up to 10% performance improvement. © 2020 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

leveraging Network Vulnerability detection using Improved Import Vector Machine and Cuckoo Search based Grey Wolf Optimizer 1

Leveraging Network Vulnerability Detection using Improved Im...

引用

1st International Conference on Optimization Techniques for learning, ICOTl 2023

作者： Preethi, P. Vasudevan, I. Saravanan, S. Krishna Prakash, R. devendhiran, A. Kongunadu College of Engineering and Technology Dept. of Computer Science and Engineering Trichy India Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Dept. of Computer Science and Engineering Chennai India Kl University Dept. of Computer Science and Engineering Vijayawada India Sona College of Technology Dept. of Information Technology Salem India Vivekanandha College of Engineering for Women Dept. of Computer Science and Technology Namakkal India

ISBN: (纸本)9798350328042

In the current interconnected and susceptible digital environment, the demand for reliable network attack prediction systems has reached a critical level. This research paper presents a novel approach for network intrusion detection utilizing the Improved Import Vector Machine (IVM) classification algorithm combined with feature selection using Cuckoo Search based Grey Wolf Optimization (CS-GWO). The aim is to improve the accuracy and effectiveness of intrusion detection systems by selecting the most relevant features and employing a powerful classification algorithm. The proposed methodology begins with data preprocessing, ensuring the dataset is appropriately cleaned and normalized. Subsequently, the CS-GWO algorithm is applied for feature selection, leveraging the strengths of both Cuckoo Search and Grey Wolf Optimization. This hybrid approach efficiently explores the search space to identify the optimal feature subset that contributes to accurate intrusion detection. Once the feature selection process is completed, the IVM classification algorithm is employed to classify network instances into intrusion or normal categories. The IVM algorithm is chosen for its ability to handle high-dimensional data and its robustness in dealing with imbalanced datasets commonly encountered in network intrusion detection. To evaluate the performance of the proposed approach, extensive experiments are conducted using benchmark datasets. Various evaluation metrics such as accuracy, precision, recall, and F1 score are utilized to assess the effectiveness of the system in detecting network intrusions. The results demonstrate that the combination of IVM classification and CS-GWO feature selection significantly improves the accuracy and efficiency of network intrusion detection. The selected features obtained through CS-GWO contribute to enhanced classification performance, enabling the system to accurately identify and respond to potential intrusions in real-time scenarios. © 2023 IE

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

detecting Human Eye Blinks through OpenCV 9

Detecting Human Eye Blinks through OpenCV

引用

9th IEEE International Conference for Convergence in Technology, I2CT 2024

作者： Wanjale, Kirti deshmukh, Araddhana Arvind Vanikar, Jyoti Chetan Adsul, Alpana Prashant Bendale, Shailesh P. Department of Computer Engineering Vishwakarma Institute of Information Technology Maharashtra Pune India Symbiosis Skill And Professional University Kiwale Pune India Symbiosis Skills And Professional University Pune Maharashtra Pune India Department of Computer Engineering Dr. D.Y. Patil College of Engineering And Innovation Affiliated To Savitribai Phule Pune University India Department of Computer Engineering Nbn Sinhgad School of Engineering Pune India

ISBN: (纸本)9798350394474

Blink detection, a crucial part of facial movement analysis, is applied in a number of fields, including human-computer interaction, driver monitoring, and healthcare. However, automatic blink recognition presents unique obstacles due to variability in blink rates and ambient conditions. This work proposes a novel real-time method to robustly and accurately detect eye blinks in video sequences using OpenCV and the Modified Eye Aspect Ratio (Modified EAR) as a crucial component. The suggested methodology's automatic facial landmark detectors are initially trained on a range of real-world datasets. When it comes to adapting to shifting surroundings, such as shifts in illumination, expressions on faces, and head orientations, these detectors are incredibly versatile. Our method carefully calculates the locations of facial landmarks with every frame in the video sequence, making it possible to calculate the vertical distance between the eyelids with accuracy. This metric, which takes the form of the Modified EAR, serves as a single scalar quantity that represents the degree of eye closure or openness. Our experimentation's empirical results highlight the facial landmarks' remarkable precision in reliably assessing the dynamics of eye opening and closure. Blinks are detected by applying a threshold value to the Modified EAR, which enables the detection of blink patterns that take place within a short temporal window. Our approach not only detects blinks efficiently but also outperforms current state-of-the-art methods, as demonstrated by the results obtained from a large and thorough dataset. © 2024 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

Face detection in Video live Stream Using deep learning 2

Face Detection in Video Live Stream Using Deep Learning

引用

2nd International Conference on disruptive Technologies, ICdT 2024

作者： Amrita Kumar, Rajiv Joshi, Shivani Raj, daksh Centre for Cyber Security and Cryptology Computer Science & Engineering Sharda School of Engineering & Technology Sharda University U.P. Greater Noida India G L Bajaj Institute of Technology and Management Department of Applied Computational Science and Engineering Nvidia Ai Learning & Research Center U.P. Greater Noida India G L Bajaj Institute of Technology and Management Department of Computer Science and Engineering U.P. Greater Noida India Galgotias College of Engineering and Technology Department of Computer Science and Engineering U.P. Greater Noida India

ISBN: (纸本)9798350371055

With the increasing popularity of live streaming platforms, there is an imperative need for robust methods for identifying individuals in real-time. The proposed system utilizes Convolutional Neural Networks (CNN) computer vision algorithms to identify faces in live streams and match them against a pre-existing database of known individuals. Additionally, the system employs deep learning and supervised learning techniques along with various types of CNN to achieve robust facial recognition in diverse lighting and environmental conditions. The system's accuracy is evaluated using various performance metrics, including precision, recall, and F1-score. The results demonstrate that the proposed system achieves high accuracy in person identification during real-time live streams. This system holds potential applications in a variety of domains, including security, entertainment, and social media. © 2024 IEEE.

关键词： computer vision Convolutional neural networks (CNN) deep learning F1-score live streaming precision recall supervised learning

来源：评论

学校读者我要写书评

暂无评论

Blockchain Innovations in Education: Empowering Students with Smart Contracts 2

Blockchain Innovations in Education: Empowering Students wit...

引用

2nd International Conference on disruptive Technologies, ICdT 2024

作者： Banu, M. Sheerin Pandian, A. Sevuga Madhavi, Bindu Sasikumar, P. Ks, Vishal Beevi, l Sherin R.M.K. Engineering College Department of Information Technology India Department of Computer Science Karnataka Bengaluru India Andhra Pradesh India Chennai India School of Commerce Vet Institute of Arts and Science Department of Commerce Tamilnadu Erode India R.M.D Engineering College Department of Computer Science and Engineering India

ISBN: (纸本)9798350371055

A lot of people are interested in how blockchain technology can be used in education because it has the potential to change everything. The main point of this study is to investigate how the Internet of Things (IoT), smart contracts, and blockchain technology can be used to give students more power in the classroom. Traditional school systems have a lot of problems when it comes to being open, safe, and effective. When old-fashioned ways of keeping records and managing data are used, wrong information, fraud, and not being able to access data are possible outcomes. In the past, though, studies have been done on separate parts of these technologies. This work suggests a complete plan to fix current issues in the education field that makes use of how blockchain technology, smart contracts, and the IoT can work together. The study uses a mixed-methods approach, which means that it uses both qualitative and quantitative evaluations. The results show that there is an effect that is good for student empowerment because it makes it easier for students to get to learning materials and generally lowers the amount of work that needs to be done. © 2024 IEEE.

关键词： Blockchain Education Technology Internet of Things Smart Contracts Student Empowerment

来源：评论

学校读者我要写书评

暂无评论

Arbitrary Kernel Embossed linear Edge Filtered Residual Network Based Sports Ball Classification 8

Arbitrary Kernel Embossed Linear Edge Filtered Residual Netw...

引用

8th IEEE International Conference on Computational System and Information Technology for Sustainable Solutions, CSITSS 2024

作者： Umanandhini, d. devi, M. Shyamala Vel Tech Rangarajan Dr. Sagunthala R&d Institute of Science and Technology Tamilnadu Chennai India Panimalar Engineering College Department of Computer Science and Engineering Tamilnadu Chennai India

ISBN: (纸本)9798331505462

Sports ball categorization serves a number of vital purposes across several kinds of business sectors that includes sports analytics, education, recreational activities, and inventory management. These objectives are successfully met by automating and improving categorization processes using cutting-edge technologies like deep learning and computer vision. This promotes creativity and efficiency in a variety of applications. With this motivation, this paper proposes Kernel Embossed Edge Filtered ResNet (KEF-ResNet) that categorizes six classes of sports ball type more effectively. The Sports Ball dataset that includes 2400 ball images that was used for implementation. Initially, the KEF-ResNet model divides ball images into six groups: football, rugby ball, baseball, cricket ball, bowling ball, and shuttlecock. during preprocessing, image labeling is done to the segregated images. Following the augmentation of data on the labelled ball images, ended with 50,400 ball images. To create Kernel Embossed Edge filtered (KEF) ball images, augmented ball images were subjected to the Kernel Embossed Edge with linear filter technique. The KEF ball images are used with both the proposed KEF-ResNet model and the current CNN techniques to assess the performance. With a high accuracy of 99.7%, the implementation shows that the suggested KEF-ResNet model performs well in the sports ball type classification task. © 2024 IEEE.

关键词： Inventory control

来源：评论

学校读者我要写书评

暂无评论

A Novel Enhanced Cuckoo Search-SVM Model for Robust Brain Tumor Segmentation 9

A Novel Enhanced Cuckoo Search-SVM Model for Robust Brain Tu...

引用

9th International Conference on Communication and Electronics Systems, ICCES 2024

作者： Alphonsa, J. Kumari, V. Sheeja Roslin dayana, K. Saveetha School of Engineering Simats University Department of Computational Intelligence Chennai600025 India R.M.D Engineering College Department of Computer Science Kavaraipettai Tiruvallur Chennai601206 India

ISBN: (纸本)9798350377972

The accurate segmentation of brain tumors from medical imaging is a critical task that aids in early diagnosis and treatment planning. Existing techniques often need help with issues like high computational demands, sensitivity to image noise, and suboptimal performance in varied imaging conditions. Enhanced Cuckoo Search-SVM (ECS-SVM) model that integrates an optimization-driven approach with machine learning for robust and efficient segmentation. We utilize the ECS algorithm to fine-tune the hyperparameters of a Support Vector Machine (SVM), which classifies extracted features from MRI images. The proposed approach leverages the search capabilities of the enhanced Cuckoo Search algorithm to identify optimal parameters that enhance the SVM's performance. With the valid application on a comprehensive dataset, the proposed method obtains a high accuracy of 95.3%. In comparative experiments, it is shown that ECS-SVM outperforms traditional and existing hybrid models, making it a great option for automated and reliable brain tumor segmentation. The results indicate significant improvements in segmentation precision, robustness, and computational efficiency, marking a notable advancement in the field. © 2024 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Algorithm-driven Optimization of Cloud Computing Architectures for Superior data Security and Efficiency 2

Algorithm-Driven Optimization of Cloud Computing Architectur...

引用

2nd IEEE International Conference on Trends in Quantum Computing and Emerging Business Technologies, TQCEBT 2024

作者： Pillai, N. Muthuvairavan Jesu Jayarin, P. david, d. Beulah Jeeva, K. Kumari, V Sheeja R.M.D Engineering College Department of Computer Science and Business Systems KavaraipettaiTamil Nadu Thiruvallur India Saveetha Institute of Medical and Technical Sciences Saveetha University Saveetha School of Engineering Department of Information Technology Tamil NaduThandalam Chennai India Care College of Engineering Department of Computer Science and Engineering Tamil Nadu Trichy India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Tamil NaduThandalam Chennai India

ISBN: (纸本)9798350384277

This paper examines how algorithm-driven optimization transforms cloud computing systems, focusing on data security and efficiency. Following a thorough cloud environment study, algorithms were picked and adapted to solve specific difficulties. After integration into the architecture, these algorithms were tested in a controlled simulation environment. CPU utilization rose from 70% to 90%, memory from 65% to 85%, storage from 60% to 80%, and network bandwidth from 75% to 95%. These improvements improve computational and storage efficiency. Intrusion detection became AI-based, and encryption increased from 128-bit to 256-bit. data breaches dropped from 5 to 1 per year, while security standards compliance increased from partial to full. These improvements strengthened cloud security. The system became more responsive from 200 to 100 ms. Server load distribution evened out, peak load handling improved, and user experience rating rose from 3.5/5 to 4.5/5. These enhancements highlighted optimized load *** consumption dropped 30%, lowering operational costs. High cooling requirements became moderate due to energy-efficient infrastructure. Sustainability goals were met by reducing annual carbon footprint by 30%. © 2024 IEEE.

关键词： Cloud security

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：