检索结果-内蒙古大学图书馆

Steel Defect Detection Using a ResNet-Inspired Model and Fourier Transforms 4

Steel Defect Detection Using a ResNet-Inspired Model and Fou...

4th International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies, ICAECT 2024

作者： Anand, Philip Sudharsan, Rohit Madhavan Ramachandran, Vishwath Prusty, Manas Ranjan Vellore Institute of Technology School of Computer Science and Engineering Chennai India Vellore Institute of Technology Centre for Cyber Physical Systems Chennai India

ISBN: (纸本)9798350343670

Defect detection in steel materials is a critical task to ensure product quality and reliability in various industrial applications. Traditional methods for detecting defects in steel surfaces are often labor-intensive, time-consuming, and prone to subjectivity and errors. However, the advent of deep learning-based models and techniques has provided a promising avenue for automating this process, significantly reducing error rates. In this research, the authors propose a Steel Defect Detection system with the help of a custom designed deep neural network inspired by the ResNet architecture. Notably, the model incorporates innovative attention layers not previously integrated into similar architectures, enhancing its predictive capabilities. The authors make use of data augmentation techniques to improve the model performance to generalize, and detect defects in imperceptible data. The novel multiclass semantic segmentation model has achieved over 91% accuracy and can be used to automate the Steel defect detection process, which can remarkably decrease the costs and time associated with manually inspecting the Steel sheets. © 2024 IEEE.

关键词： Attention layers augmentation Deep learning Residual Network Steel Defect

来源：评论

学校读者我要写书评

暂无评论

Enhanced Research on Legal Data Extraction and Document Analysis Tool 8

Enhanced Research on Legal Data Extraction and Document Anal...

引用

8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud), I-SMAC 2024

作者： Gupta, Devansh Bhoji, Spoorthi Arora, Maadhav Suganya, R. Geetha, S. Manjula, D. Vellore Institute of Technology School of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350376425

Certainly, in the practice of law, the issue of case law and legal document mining and understanding is at the forefront among the priorities of both practicing lawyers and students. For this reason, this research proposes an approach to developing a solution that employs the most current web scraping and natural language processing principles to address this problem. By using Selenium, BeautifulSoup, and Requests data from famous legal websites such as Indian Kanoon and Advocate Khoji are collected and saved on the hard disk in CSV format systematically. Subsequently, the users can choose specific datasets and ask questions concerning the content. To assist the user in understanding upcoming events and to provide semantically meaningful and contextually correct answers, the system connects API calls with two professional language models, which are Gemini and Lama 7B-Instruct. The entire workflow is being implemented by the Streamlit module in Python That's how the user interface is realized. The purpose of this tool is to cut down the time and work that it takes to work on legal research to the barest minimum by increasing the efficiency of the tool and improving the experience of the lawyers and law students. © 2024 IEEE.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Security Enhancement in Social Media Users using Keystroke Biometrics 15

Security Enhancement in Social Media Users using Keystroke B...

引用

15th International Conference on Advances in Computing, Control, and Telecommunication Technologies, ACT 2024

作者： Kavitha, D. Thejas, S. School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Chennai India

ISBN: (纸本)9798331300579

As social media platforms continue to grow in popularity, ensuring the security and privacy of users' accounts has become a critical *** authentication methods, such as passwords and PINs, are vulnerable to various attacks, including password guessing, phishing, and brute force *** address these challenges, researchers have turned to keystroke dynamics as a promising *** dynamics, also known as keystroke biometrics, leverages the unique typing patterns and behavioral characteristics of individuals to authenticate their *** approach analyzes various keystroke attributes, such as key press duration, flight time, and typing rhythm, to create a user's unique typing *** continuously monitoring and comparing these patterns, social media platforms can verify the authenticity of users in *** paper explores the potential of keystroke dynamics in enhancing security for social media *** discusses the advantages of this approach, including its non-intrusive nature, continuous authentication capabilities, and resistance to common ***, the paper examines the challenges associated with implementing keystroke dynamics, such as user acceptance, system performance, and potential privacy concerns. © Grenze Scientific Society, 2024.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

Transformative Approaches in Landslide Detection-Synergizing Cutting-Edge Deep Learning Models with Optimal Fusion Strategies

Transformative Approaches in Landslide Detection-Synergizing...

引用

2024 IEEE Silchar Subsection Annual Conference, SILCON 2024

作者： Saimirra, R. Srivats, R. Johnson, Deepika Roselind Siddiqui, Muskaan Sharma, Abhiram School of Computer Science and Engineering Vellore Institute of Technology Chennai Campus Chennai India

ISBN: (纸本)9798331540821

Landslides are catastrophic natural disasters causing significant loss of life, displacement, and economic damage globally. To enhance landslide detection accuracy, an optimal methodology using advanced deep learning models has been developed. Five state-of-the-art models- VGG16, DenseNet201, ResNet50, InceptionV3, and a Generative Adversarial Network (GAN), were trained and evaluated. VGG16 and DenseNet201, the top-performing models, were integrated through feature fusion and global average pooling to combine their strengths. Data augmentation techniques were used to improve model robustness, and hyper-parameters were fine-tuned with a reduced learning rate for optimal performance. The combined model achieved 97.22% accuracy, 97.78% precision, 96.70% recall, and a 97.24% F1-score. Performance was demonstrated with detailed training and validation accuracy and loss curves, a confusion matrix, and an ROC curve with an AUC of 0.96. This methodology significantly enhances the reliability of emergency response efforts, particularly in disrupted communication networks, enabling timely interventions. A comparative analysis highlighted the superiority of this integrated approach over individual models. This study presents a novel pipeline for landslide detection, achieving advanced recognition through innovative model integration and optimization, contributing to effective disaster mitigation and long-term risk management. © 2024 IEEE.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Malicious URL Detection: Regular Expression Analysis, Machine Learning, and VirusTotal API 2nd

A Comparative Study of Malicious URL Detection: Regular Expr...

引用

2nd International Congress of Electrical and computer engineering, ICECENG 2023

作者： Misquitta, Jason Kannan, Anusha School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Chennai India

ISBN: (纸本)9783031527593

In this paper, malicious URL detection was done in three ways. In the first method, self-written code was used that breaks up the URL into several fragments and then using functions and checkers, it analyzes whether the URL is malicious. The second approach involved utilizing a fixed csv dataset containing website URLs from over 11,000 websites. Each entry in the dataset included 30 parameters describing the website and a class label indicating whether it was classified as a phishing website (1) or not (−1). Nine machine learning models were compared on this dataset to see which models gave the highest accuracy and F1-score. The last method was done using VirusTotal API Key. The user has to input the URL of a website and the code classifies it as malicious or not. A report is also generated if the URL is detected as malicious. The report is a compilation of the analysis of that website by several security vendors. The results showed that the self-written code achieved a 90% accuracy in detecting malicious URLs. All the nine machine learning models except for Naive Bayes Classifier obtained extremely high values for accuracy, F1-score, and recall. Thus, our study contributed to cybersecurity by demonstrating the effectiveness of multiple malicious URL detection methods, including a self-written code, machine learning models, and VirusTotal API, providing insights into their accuracy and potential applications. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Websites

来源：评论

学校读者我要写书评

暂无评论

Classification of Skin Lesions to Predict Melanoma using Deep Learning Algorithm on ESP32 2

Classification of Skin Lesions to Predict Melanoma using Dee...

引用

2nd International Conference on Computational Intelligence and Network Systems, CINS 2024

作者： Saranyaraj, D. Parvathi, R. Pattabiraman, V. School of Computer Science and Engineering Vellore Institute of Technology Chennai Campus Chennai India

ISBN: (数字)9798331504106

ISBN: (纸本)9798331504106

This work describes a computer image-based deep learning system for identifying and classifying skin cancer that is developed using an ESP-32 microcontroller. We train and test our CNN model using a dataset of over 10,000 images obtained from HAM10000. This research classifies skin cancer into seven kinds. This research addresses the most prominent skin cancer types as classes such as melanoma, basal cell carcinoma, benign keratosis, dermatofibroma, nevus, actinic keratosis, and vascular lesions. CNN algorithm is used to perform the detection which is known for its feature extraction and ease the use of large data which in this case is various types of skin cancer which is processed in the training model. In addition to CNN, we suggested a novel approach that uses ViT-B16, ViT-B32, EfficientNet V2, ResNet-50 V2, and ensemble methods to enhance prediction results on skin cancer diagnosis. We suggest using ViT-B16 and ViT-B32 in combination with EfficientNet V2 and ResNet-50 V2, a cutting-edge deep learning architecture, that extracts features from skin lesion images. To increase the classification model’s performance, three distinct ensemble approaches are used: weighted average ensemble, geometric mean ensemble, and average ensemble. The collected characteristics are then loaded into a binary classification model, which uses all of the approaches discussed above to discriminate between melanoma and benign skin lesions. Our CNN model surpassed prior recent techniques, achieving an accuracy of 83 % percent when deployed on ESP-32. Our findings show that deep learning-based approaches applied to microcontrollers can reduce computing costs and develop a portable/trained device that predicts skin cancer based on test data. ©2024 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Performance Improvement through Novel Adaptive Node and Container Aware Scheduler with Resource Availability Control in Hadoop YARN

引用

computer Systems science & engineering 2023年第12期47卷 3083-3108页

作者： J.S.Manjaly T.Subbulakshmi School of Computer Science and Engineering Vellore Institute of TechnologyChennai600127India

The default scheduler of Apache Hadoop demonstrates operational inefficiencies when connecting external sources and processing transformation *** paper has proposed a novel scheduler for enhancement of the performance of the Hadoop Yet Another Resource Negotiator(YARN)scheduler,called the Adaptive Node and Container Aware Scheduler(ANACRAC),that aligns cluster resources to the demands of the applications in the real *** approach performs to leverage the user-provided configurations as a unique design to apportion nodes,or containers within the nodes,to application ***,it provides the flexibility to the applications for selecting and choosing which node’s resources they want to manage and adds limits to prevent threshold breaches by adding additional jobs as *** or container awareness can be utilized individually or in combination to increase *** top of this,the resource availability within the node and containers can also be *** paper also focuses on the elasticity of the containers and self-adaptiveness depending on the job *** results proved that 15%–20%performance improvement was achieved compared with the node and container awareness feature of the *** has been validated that this ANACRAC scheduler demonstrates a 70%–90%performance improvement compared with the default Fair *** results also demonstrated the success of the enhancement and a performance improvement in the range of 60%to 200%when applications were connected with external interfaces and high workloads.

关键词： Big data Hadoop YARN hadoop distributed file system(HDFS) MapReduce scheduling fair scheduler

来源：评论

学校读者我要写书评

暂无评论

Multi-stage deep investigation pipeline on detecting malign network traffic

引用

Materials Today: Proceedings 2022年 62卷 4726-4731页

作者： Nasreen Fathima, A.H. Syed Ibrahim, S.P. School of Computer Science and Engineering Vellore Institute of Technology Chennai India

To detect intrusions effectively, we need vulnerable information. To collect the network traffic transactions and assess the anomaly which includes essential details related to network traffic data, we explore the datasets that have intrusion related network traffic. Based on this assessment, to predict attacks in the network transaction efficiently, we use the most upgraded benchmark datasets NSL-KDD and CIC-IDS-2017. In this work, a novel multistage deep learning model is developed to evaluate the optimal features for training. Our proposed model identifies weird patterns of network traffic with probabilistic unsupervised learning technique variational autoencoder. By pre-training the model in the first and second stage to classify the attack with fusion of random forest at the final stage, we improve the outcome of a very biased class dataset. Experimental results unveil the improvisation of the overall computational and time complexity. The proposed model accuracy measures as 99% with improved precision for CIC-IDS-2017 dataset. Class imbalance handling techniques have to be combined as future work in order to reduce the high false positive rates and accordingly increase the low false negative rates for the minority classes. © 2022

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

A Comprehensive Review of Air Quality Prediction Techniques: Methods, Performance and Challenges 8

A Comprehensive Review of Air Quality Prediction Techniques:...

引用

8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud), I-SMAC 2024

作者： Sellamuthu, Suseela Niranjan Ayush Manori, Vaidik Chaurasia, Yash Mohan, Prakarsh Vellore Institute of Technology School of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350376425

Air quality is what makes the places we live in habitable and affects the well-being and productivity of the people. Not paying attention to air quality can be dangerous, which has raised alarm among scientists, researchers, and the general public because it has been getting worse over the past decades. This paper will look at various methods and techniques for monitoring and predicting Air Quality Indexes (AQI) and how complicated it is to predict air quality exactly. Diverse machine learning models have been developed as well as data-driven approaches to improve the accuracy and credibility of air pollution forecasts. The dynamic nature of environmental conditions may explain why despite advances, predicting air quality remains a difficult task due to many interfering variables. This paper also proposes an approach for the creation of an efficient alert system. By using current data that offers predictive models that should help with time-sensitive real-time alerts about poor air quality to minimize any negative effects from such occurrences, these predictions should provide up-to-date results on air pollution indices. Such an alert system could be integrated with existing operational techniques, which would improve air quality management and as a result, the safety of life and productivity of individuals. © 2024 IEEE.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Neural Network-Based Alzheimer's Disease Diagnosis: A Deep Learning Approach

Neural Network-Based Alzheimer's Disease Diagnosis: A Deep L...

引用

2024 IEEE Students Conference on engineering and Systems, SCES 2024

作者： Shri Varshan, P. Suganya, R. School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Chennai India

ISBN: (纸本)9798350374711

This research presents a novel hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model technique for Alzheimer's disease (AD) identification. By utilizing data from structural and functional neuroimaging, the suggested model seeks to increase the precision and resilience of AD diagnosis. By extracting spatial features from structural MRI scans with CNNs and capturing temporal patterns from functional MRI (fMRI) time-series data with LSTMs, the model demonstrates superior performance compared to traditional methods and standalone CNN or LSTM models. Evaluation on a large dataset showcases high accuracy, sensitivity, and specificity in classifying AD patients and healthy controls. Moreover, interpretability analysis of the model's learned features and attention mechanisms offers insights into AD biomarkers, facilitating a better understanding of the disease's pathology. Overall, the CNN-LSTM hybrid model presents a promising avenue for early AD detection, with implications for enhancing clinical diagnosis and treatment strategies. As a result of training the model along with other models, the CNN-LSTM model showed us a promising validation accuracy of 92.13%, which shows us the effectiveness of the model. © 2024 IEEE.

关键词： Convolutional neural 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：