检索结果-内蒙古大学图书馆

2nd International Conference on Edge Computing and applications, ICECAA 2023

作者： Agrawal, Subhash Chand Tripathi, Rajesh Kumar Bhardwaj, Neeraj Parashar, Prashun Gla University Department of Computer Engineering & Applications Mathura India

ISBN: (纸本)9798350347579

Air paint application is a technology that has gained popularity due to its ability to simulate real-life painting tasks in a virtual environment. This technology allows users to experiment with different colors, textures, and finishes without the need for physical paint or equipment. In this study, a solution is proposed that allows to draw anything virtually using a camera and a colored marker. The marker is usually placed on the tip of the finger and its movement is recorded by the camera. computer vision techniques are used for the solution to this problem with the help of its extensive libraries, simple syntax, and ease of use. However, this problem can also be implemented in other similar open cv supported languages with some basic understanding. This is achieved by tracking and detecting the color of the marker. Once the color is recognized, a mask is created. Morphological operations is performed such as erosion and dilation on the mask. Erosion reduces the impurities in the mask while dilation restores the main mask that has been eroded. The aim of this study is to allow us to draw virtually without the need for physical drawing tools. One of the major applications of this study is to improve the teaching learning process. An instructor with the help of virtual drawing can create effective contents for his/her class like can draw different shapes, tables and write text, can create flowchart, diagram, etc. © 2023 IEEE.

关键词： Python

来源：评论

学校读者我要写书评

暂无评论

IoV-fog-cloud framework for road anomalies detection using SVM-nAVDD approach

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第8期14卷 10899-10915页

作者： Kumar, Navin Sood, Sandeep Saini, Munish Department of Computer Engineering and Technology Guru Nanak Dev University Punjab Amritsar India Department of Computer Applications National Institute of Technology Haryana Kurukshetra India

The detection of road anomalies in highway management is a crucial yet manpower-demanding task. Although there have been many previous types of research based on visual and acceleration data collected from vehicles. However, to distinctively differentiate a pothole and speedbump from acceleration remains a challenge. The visual methods require high computation supported by expensive GPUs. This paper proposes a novel contribution to depth detection using angular velocity for pothole detection. The solution is accorded by three-layer architecture inclusive of fog-based cloud infrastructure where each layer performs a respective function based on computing capability. The data acquisition is accomplished using an IoV setup equipped with M8N GPS, and MPU-6050 mounted on an Arduino MEGA 2560. The results are compared with the state-of-art algorithms and methodologies. The proposed SVM-nAVDD achieved the precision of 0.9092–0.9123 for the tested datasets and depth/height detection with 92% accuracy. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Data acquisition

来源：评论

学校读者我要写书评

暂无评论

Predicting Risky Environment for Child Inside House using Deep Learning 5

Predicting Risky Environment for Child Inside House using De...

引用

5th International Conference on Emerging Smart Computing and Informatics, ESCI 2023

作者： Ahmad, Naeem Arya, Sumedha Singh, Deepak Department of Computer Applications Raipur India Department of Computer Science and Engineering Raipur India

ISBN: (纸本)9781665475242

We place a high priority on making sure kids are safe. Yet in the modern world, there are a lot of threats to children's safety and challenges to achieving other safety goals. Parent can't always supervise their kids and leave them alone inside the house. So this is a big problem of our time as the small kids is unable to assess the risk due to any harmful activity or objects. The purpose of this study is to provide safety to the child by alerting the parent about the child's position and surroundings. This proposed model can be used to monitor child's movements. We proposed a children's video data set. The video data set contains every moment of the child in the form of frames, and 30 frames are created every second. Additionally, we train a baseline computer vision module that determines the level of dangerousness for a child using deep learning. The baseline consists of well-established methods like conditional and logistic regression for child's pose estimation, as well as cutting-edge deep learning image classifiers. We then integrate the stumbled-condition baseline into various learning platforms and evaluate performance in terms of accuracy and loss. © 2023 IEEE.

关键词： computer vision

来源：评论

学校读者我要写书评

暂无评论

SLDERCLC: Improving Smart Learning Frameworks Through Deep Learning Enhanced Recommender Systems Using an Efficient Cross-Layer Collaborative Approach

引用

SN computer Science 2025年第2期6卷 1-18页

作者： Kaur, Harjinder Kaur, Tarandeep Kumar, Mukesh Bhardwaj, Vivek School of Computer Applications Lovely Professional University Punjab Phagwara 144411 India Department of Computer Applications Chandigarh School of Business Chandigarh Group of Colleges Jhanjeri Punjab Mohali 140307 India School of Computer Science and Engineering Manipal University Jaipur Jaipur 303007 India

The exponentially scaling domain of a smart learning frameworks necessitates advanced recommender systems capable of providing precise and relevant educational content to learners. Traditional recommender systems often grapple with challenges such as data sparsity, cold start problems, and the inability to capture complex user-item interactions & scenarios. This work introduces a novel approach to recommender systems in smart learning frameworks, addressing these limitations by leveraging deep learning techniques. Our proposed model amalgamates Deep Q Learning with Multilayer Graph Neural Networks (GNNs) for Cross Layer Collaborative Filtering and integrates Autoencoders with Capsule Networks for enhanced recommendation accuracy levels. The use of Deep Q Learning facilitates efficient decision-making in dynamic environments, while Multilayer GNNs adeptly handle relational data, improving the recommendation process by capturing the intricate connections within the educational contents. Furthermore, the incorporation of Autoencoders with Capsule Networks offers a sophisticated mechanism for understanding hierarchical relationships in data, which is crucial for personalized learning paths. The effectiveness of our model is substantiated through rigorous testing on the ASSISTments and EdNet datasets. The results are compelling, showcasing a 4.9% increase in precision, 5.5% improvement in accuracy, 3.5% higher recall, 2.9% greater AUC (area under the curve), 3.4% increased specificity, and a notable 4.5% reduction in delay for smart learning recommendations in comparison with SBBR, DRL and ROME. These improvements highlight the model’s proficiency in delivering timely and relevant educational content, thereby enhancing the learning experience of students as well as the teaching strategies of faculty. The framework attains significant improvements in the efficiency and effectiveness of educational content recommendations thereby increasing the retention rate of students w

关键词： Autoencoders Deep learning Graph Neural Networks Recommender systems Smart learning

来源：评论

学校读者我要写书评

暂无评论

Evaluating Blockchain Platforms Using an Integrated Multi-Criteria Decision-Making Approach: AHP and TOPSIS Methods

Evaluating Blockchain Platforms Using an Integrated Multi-Cr...

引用

2024 IEEE Region 10 Conference, TENCON 2024

作者： Bhol, Seema Gupta Kumar, Sahoo Sanjib School of Computer Applications Kiit University Bhubaneswar India National University of Singapore Department of Electrical and Computer Engineering Singapore

ISBN: (纸本)9798350350821

Blockchain technology is gaining immense popularity in the past few years, and is extensively applied in various domains. The diversity of commercial blockchain platforms and their unique features poses a challenge in selecting optimal platform for meeting requirements of an organization. For successful implementation of block chain based application, selection of appropriate blockchain platform is very crucial. The present study proposes a methodology to select the best blockchain platform using a Multi-Criteria Decision-Making (MCDM) approach. An Integrated approach of Analytic Hierarchy Process (AHP) and TOPSIS method is adopted for ranking of blockchain platforms. Five blockchain platforms were evaluated on the basis of seven criteria namely: transaction fees, security, transaction speed, interoperability, scalability, energy requirements, and smart contracts as use case. The utility of the proposed approach has been established through an illustrative example for ranking blockchain platforms. The results demonstrate how and integration of AHP and TOPSIS methods can effectively prioritize Blockchain platforms based on specified criteria while reducing inconsistencies, ensuring the best possible decision in blockchain platform selection. © 2024 IEEE.

关键词： Smart contract

来源：评论

学校读者我要写书评

暂无评论

Optimal Selection of Machine Learning as a Service (MLaaS) Using a Hybrid Multi-Criteria Decision Making Approach: Integrating Entropy Weight and VIKOR Methods

Optimal Selection of Machine Learning as a Service (MLaaS) U...

引用

2024 IEEE Region 10 Conference, TENCON 2024

ISBN: (纸本)9798350350821

Cloud computing has change the landscape of IT industry. Apart from traditional Software a Service (SaaS) and Product as Service (PasS), a relatively new service called Machine Learning as a Service (MLaaS) is being offered by various cloud service providers. IT professionals, data Scientists and other users can benefit from pre-built machine learning algorithms, libraries, and frameworks, as well as access to large-scale computing resources and data storage. MLaaS offers a user friendly interface that assists in development and deployment of machine learning models for various applications. MLaaS landscape is continually evolving, with new features and offerings being added regularly. Its importance is likely to continue growing as more industries and sectors embrace the benefits it offers. Many commercial MLaaS Cloud service providers are offering tools and techniques under a big umbrella called Machine Learning over a cloud as service. However, selecting the best MLaaS service provider is crucial to meet business needs of an organization. The present paper is proposing Multi-criteria Decision Making (MCDM) based approach to select best MLaaS providers. Hybrid application of Entropy method and VIKOR method is used where Entropy method is used for calculation of criteria weights and VIKOR method used these weights for final ranking of alternative MLaaS providers. © 2024 IEEE.

关键词： Software as a service (SaaS)

来源：评论

学校读者我要写书评

暂无评论

Unveiling factuality and injecting knowledge for LLMs via reinforcement learning and data proportion

引用

Science China(Information Sciences) 2024年第10期67卷 389-390页

作者： Wenjun KE Ziyu SHANG Zhizhao LUO Peng WANG Yikai GUO Qi LIU Yuxuan CHEN School of Computer Science and Engineering Southeast University Key Laboratory of New Generation Artificial Intelligence Technology and Its Interdisciplinary Applications(Southeast University) Beijing Institute of Technology Zhuhai Beijing Institute of Computer Technology and Application

Large language models(LLMs) have demonstrated remarkable effectiveness across various natural language processing(NLP) tasks, as evidenced by recent studies [1, 2]. However, these models often produce responses that conflict with reality due to the unreliable distribution of facts within their training data, which is particularly critical for applications requiring high credibility and accuracy [3].

关键词：

来源：评论

学校读者我要写书评

暂无评论

Empowering Early Detection: CervicalCareXG Revolutionizes Cervical type cancer Diagnostics 4

Empowering Early Detection: CervicalCareXG Revolutionizes Ce...

引用

4th IEEE Asian Conference on Innovation in Technology, ASIANCON 2024

作者： Verma, Shanti Suthar, Hiral Rami, Khyati L J University Department of Computer Applications Gujarat Ahmedabad India Silver Oak University Department of Computer Applications Gujarat India Sal Institute of Technology & Engineering Research Department of Computer Science Gujarat India

ISBN: (纸本)9798350354218

A XGBoost-Based System in Cervical type cancer Diagnostics are proposed in the current research work. Generally Cervical type of cancer will be a significant health concern worldwide, particularly in the regions with limited access on advanced healthcare facilities. On Early detection will plays a pivotal role on improving patient outcomes and by reducing mortality rates. In this study, the propose system termed CervicalCareXG, a novelistic system for cervical type cancer diagnostics depends on the XGBoost methodology. CervicalCareXG will utilizes a machine Learning-Driven approach to analyze patient data, also including demographic information, medical history, and cervical screening results, to Prediction the likelihood of cervical type cancer development. The XGBoost algorithm, known for its prediction accuracy and efficacy in handling complex health related datasets, is employed to train a predictive model with a capable of identifying high-risk individuals who may requires further diagnostic evaluation or preventive interventions. Through comprehensively evaluation using reality patient data, CervicalCareXG may demonstrates a promising performance in terms of sensitivity, specificity, and area under the receivers operating characteristic curve (AUC-ROC). Thus, it showcases its potential as a valuable tool for healthcare professionals on cervical type cancer screening programs. By harnessing the power of machine Learning-Driven and XGBoost, CervicalCareXG with an aim to contribute to the early detection and management of cervical type cancer, ultimately by improving patient outcomes and reducing the burden of this devastating disease. © 2024 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Single Under-Water Image Enhancement Using the Modified Transmission Map and Background Light Estimation

引用

International Conference on Machine Vision and Augmented Intelligence, MAI 2022

作者： Verma, Gunjan Kumar, Manoj Raikwar, Suresh Department of Computer Engineering and Applications GLA University Uttar Pradesh Mathura281406 India Department of Computer Engineering and Applications Thappar University Punjab Patiala147004 India

ISBN: (纸本)9789819901883

In under-water environment, sensors are used to explore marine resources, sea-bed, archaeology, and tracking. Due to that, vision sensor plays a vital role as it helps to obtain high-content information. The light attenuation and backscattering in under-water environment are dependent on the wavelength that leads to color distortion, low contrast, and blurry images. The under-water image formation model depends on mainly two factors: background light and transmission light. In this paper, a modified transmission light estimation method and a new attenuation coefficient (U-TMBL) are proposed that enhance the quality of under-water images. Also, novel background light estimation method is proposed. The performance of the proposed method is quantitatively and qualitatively analyzed. The quantitative comparison is based on NIQE, BRISQUE, and UIQM evaluation metrics. The quantitative and qualitative results demonstrate that the proposed U-TMBL overcomes the existing under-water image enhancement approaches. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Light transmission

来源：评论

学校读者我要写书评

暂无评论

Unmasking Hypertension Through a Fusion of Human Lifestyle Data using Machine Learning Algorithms 6

Unmasking Hypertension Through a Fusion of Human Lifestyle D...

引用

6th International Conference on Mobile Computing and Sustainable Informatics, ICMCSI 2025

作者： Vishnu Priya, V. Nisanth, G. Nivethaa, R.P. Vijay, Srinidhi Department of Computer Science and Engineering Kongu Engineering College Erode Perundurai India Department of Computer Applications Kongu Engineering College Erode Perundurai India Department of Computer Technology UG Kongu Engineering College Erode Perundurai India

ISBN: (纸本)9798331522667

Hypertension, also referred to as high blood pressure, is a condition arising from the consistently high blood pressure against artery walls. The volume and output of blood from the heart primarily control blood pressure under stiff or tight arterial conditions. High pulse pressure leads to increased blood pressure, since the pumping of more blood makes the arteries inactive. The World Health Organization has a comprehensive guideline in which an evidence-based pharmacological approach has been recommended for managing adult patients with hypertension, not neglecting to mention lifestyle modification in order to support medicinal treatment. Machine learning may revolutionize the way hypertension treatment is projected for the management of personalized treatment routines for patients based on their characteristics. ML provides great promise in seeking ideal medication and predicting responses thereof against massive sets of data for hypertensive patients. The techniques allow for the determination of the individual risk for hypertension and ensure that treatment protocols are optimal. Compared to methods used to predict hypertension in the past, ML algorithms show a marked improvement in performance. © 2025 IEEE.

关键词： Personalized medicine

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：