检索结果-内蒙古大学图书馆

Advanced Computing & Communication Technologies (ICACCTech), International Conference on

作者： Parth Middha Syed Basim Muitaba Andrabi Tanushree Gupta Abhishek Singhal Dept. of Computer Science & Engineering Amity University Uttar Pradesh Noida India

ISBN: (数字)9798331519056

ISBN: (纸本)9798331519063

Ultraviolet radiation from sunlight has both beneficial and harmful effects on human health, with the UV index serving as a crucial metric for assessing potential harm. Excessive exposure can lead to adverse effects like skin cancers, cataracts, and immune suppression. Air quality influences the UV index by affecting the scattering and absorption of UV radiation, with particulate matter and pollutants altering the amount of radiation reaching the Earth's surface. Recognizing the dynamic relationship between air quality and the UV index is vital for understanding their combined impact on human health and the environment. This work utilizes the real-time weather information from prominent cities across the world, capturing diverse climatic conditions and air quality variations. With a particular emphasis on UV index prediction, the study focuses on employing a comparative analysis of various regression models, including linear regression, decision trees, support vector machines, logistic regression, random forest, XGBoost, LightGBM, gradient boosting and neural networks. The comparative analysis aims to identify the most proficient model for capturing the intricate relationships between chosen meteorological variables, including temperature, wind degree, precipitation, and others, in relation to the UV index. This paper explores the impact of air quality, temperature, wind degree, precipitation, and other relevant factors on UV index variations, offering insights into the complex interplay between meteorological parameters and ultraviolet radiation. The findings of this research hold potential implications for public health, as understanding the factors influencing UV index can contribute to enhanced sun exposure risk assessments. Furthermore, the comparative analysis of regression models provides valuable guidance for future studies aiming to predict UV index variations based on real-time weather data, thereby contributing to the development of more accurate and regi

关键词： Analytical models Wind Temperature Atmospheric modeling Urban areas Predictive models Air quality Real-time systems Indexes Public healthcare

来源：评论

学校读者我要写书评

暂无评论

Data Augmentation's Effect on Machine Learning Models when Learning with Imbalanced Data

Data Augmentation's Effect on Machine Learning Models when L...

引用

International Conference on Data science and Advanced Analytics (DSAA)

作者： Damien A. Dablain Nitesh V. Chawla Dept. Computer Science and Engineering University of Notre Dame Notre Dame USA

ISBN: (数字)9798350364941

ISBN: (纸本)9798350364958

Real-world data is often imbalanced, such that the number of training instances varies by class. Data augmentation (DA) of under-represented classes is commonly used to improve model generalization in the face of class imbalance. Despite its ubiquity, the impact of data augmentation on machine learning (ML) models is not clearly understood. Here, we undertake a holistic examination of the effect of DA on under-represented classes. Unlike other studies, which focus on a single ML model type, we examine three different classifier families: convolutional neural networks, support vector machines, and logistic regression models; five different DA techniques and two different data modalities - image and tabular. Our research indicates that DA, when applied to imbalanced data, produces substantial changes in model weights, support vectors and front-end feature selection. These changes occur with respect to all classes, not just the ones that DA is applied to. Further, our empirical analysis shows that data augmentation's positive influence on generalization does not necessarily occur as a result of reducing weight norms. Rather, weight and support vector specialization play important roles in generalization. The specialization process may be a form of memorization that is spawned by variances introduced by augmented data. We investigate the seeming contradiction between improved generalization versus weight and support vector specialization.

关键词： Training Support vector machines Logistic regression Neural networks Machine learning Data augmentation Feature extraction Data models Vectors Faces

来源：评论

学校读者我要写书评

暂无评论

Plant Disease Detection using Convolutional Neural Network

Plant Disease Detection using Convolutional Neural Network

引用

Smart Devices (ICSD), International Conference on

作者： Jyoti Agarwal Shivansh Naithani Utkarsh Singh Harsh Raj Dept. of Computer Science and Engineering Graphic Era Deemed to be University Dehradun India

ISBN: (数字)9798350394412

ISBN: (纸本)9798350394429

Pest-infested plants and crops have a negative effect on the nation's agricultural output. To detect and identify diseases in plants, farmers or other experts must keep a careful watch on plants. However, this method is frequently expensive, time-consuming, and inaccurate. Finding a spot on the leaves of the infected plant is one method of detecting plant diseases. This research aims to develop a disease detection model based on the classification of leaf images. To detect plant illnesses the proposed model utilizes image processing with a convolution neural network (CNN). This research work focus on three different plant diseases: tomato, corn and potato. Experimental results show high accuracy rate of 98.8%. Proposed work is also compared with some existing work working o same plants diseases.

关键词： Plant diseases Convolution Plants (biology) Image processing Neural networks Crops Forestry Vectors Convolutional neural networks Smart devices

来源：评论

学校读者我要写书评

暂无评论

Water Region Extraction in Aerial Images Using Deep Learning

Water Region Extraction in Aerial Images Using Deep Learning

引用

Intelligent Computing and Sustainable Innovations in Technology (IC-SIT), International Conference on

作者： Nikita Pandey Saumya Kumar Satish Kumar Yadav Shashwat Srivastava Shomil Singh Deepak Gupta Dept. of Computer Science and Engineering Pranveer Singh Institute of Technology Kanpur India

ISBN: (数字)9798350369175

ISBN: (纸本)9798350369182

Water bodies are essential component of Earth’s surface and accurate water region mapping is crucial for disaster management and environmental monitoring. Existing methods to map water bodies relies on traditional modelling methods that need extensive manual calibration struggle with scalability and lack advanced pattern recognition. This paper evaluates the effectiveness of deep learning-based semantic segmentation models, DeepLabV3+ and U-Net, using aerial images resized to $\mathbf{2 5 6 x} 256$ pixels in mapping water regions in aerial images. DeepLabV3+, with a ResNet50 backbone, and U-Net are trained on a comprehensive flood-affected dataset with preprocessing techniques such as resizing and augmentation. Model performance is assessed using metrics like IoU, F1 score, accuracy, precision, and recall. The results indicate that DeepLabV3+ outperforms U-Net, showing promise for automating water region mapping and improving flood surveys and urban planning.

关键词： Training Measurement Deep learning Surveys Technological innovation Accuracy Computational modeling Semantic segmentation Urban planning Water resources

来源：评论

学校读者我要写书评

暂无评论

Adaptive Traffic Signal Management Using Real-Time Vehicle Detection and Tracking

Adaptive Traffic Signal Management Using Real-Time Vehicle D...

引用

Interdisciplinary Approaches in Technology and Management for Social Innovation (IATMSI), IEEE Conference on

作者： Sarvesh Shirulkar Rishabhdev Makode Richa Khandelwal Dept. of Electronics and Computer Science Shri Ramdeobaba College of Engineering and Management Nagpur India Dept. of Electronics Engineering Ramdeobaba University Nagpur India

ISBN: (数字)9798331521691

ISBN: (纸本)9798331521707

Traffic congestion is a significant issue in urban cities, as it restricts traffic flow, increases fuel consumption, and produces high emissions. Traditional traffic signal systems work on fixed timing without considering real-time variations in traffic and often worsen the problem. This paper describes an adaptive traffic signal control system based on the YOLO (You Only Look Once) algorithm, using object detection for real-time signal timing in response to real-time traffic. The system applies real-time vehicle detection and tracking at intersections for dynamic management of green light time according to volume and vehicle type. The proposed approach makes vehicle recognition significantly efficient and responds quickly to the change in traffic situation. The paper discusses the design, implementation, and evaluation of performance in a simulated environment. The results demonstrate improved traffic management and reduced congestion, offering a promising solution to smart city traffic control.

关键词： YOLO Adaptive systems Smart cities Vehicle detection Green products Traffic control Real-time systems Timing Dynamical systems Vehicle dynamics

来源：评论

学校读者我要写书评

暂无评论

Vyoma-ADS : An AI-Driven Tool for Adaptive Monitoring and Anomaly Detection for Space Agriculture

Vyoma-ADS : An AI-Driven Tool for Adaptive Monitoring and An...

引用

Communication, Computing and Signal Processing (IICCCS), IEEE International Conference on

作者： Rajdeep Das Vaishnavi Sharma Priyanka Kumari Abhishek Raj Sudipta Das Indrajit De Dept. of Computer Science and Business Systems Institute of Engineering and Management Kolkata India

ISBN: (数字)9798350390759

ISBN: (纸本)9798350390766

As humanity ventures further into space exploration, securing sustainable resources becomes paramount. The success of space farming depends not only on the ability to cultivate crops but also on the effective management of environmental conditions. Introducing Vyoma-ADS: an innovative system that integrates IoT technology with an adaptive Anomaly Detection System, specifically designed for space agriculture. Our study underscores the crucial role of environmental monitoring within space shuttles, where even minor fluctuations can have a significant impact on plant growth and productivity. To tackle these challenges, we designed a compact IoT-enabled system to monitor the moisture content in the soil, the temperature and humidity of the surrounding environment, and the ppm value of unwanted gases in the atmosphere. Through rigorous testing and analysis, we established optimal thresholds conducive to plant growth in the unique conditions of space. Recognizing the inherent variability and occasional anomalies in environmental data, we investigated three distinct machine learning models for anomaly detection: Isolation Forest, One-Class Support Vector Machines (OCSVM) and Density-Based Spatial Clustering of applications with Noise (DBSCAN). Our findings reveal that DBSCAN outperforms other techniques in identifying anomalies amidst fluctuating data. Additionally, we detail our approach to data visualization using a SaaS cloud platform, which enhances the comprehensibility of the study and facilitates better decision-making processes. This paper demonstrates how the integration of IoT monitoring and advanced anomaly detection methods can significantly improve the reliability and efficiency of space agriculture. By leveraging these technologies, we aim to create a robust system capable of maintaining optimal growing conditions for plants in space.

关键词： Temperature sensors Cloud computing Atmospheric modeling Space shuttles Data visualization Machine learning Humidity Agriculture Sensors Anomaly detection

来源：评论

学校读者我要写书评

暂无评论

A Machine Learning-Based Secure Face Verification Scheme and Its applications to Digital Surveillance

arXiv

引用

arXiv 2024年

作者： Wang, Huan-Chih Wu, Ja-Ling Dept. Computer Science and Information Engineering National Taiwan University Taipei Taiwan

Face verification is a well-known image analysis application and is widely used to recognize individuals in contemporary society. However, most real-world recognition systems ignore the importance of protecting the identity-sensitive facial images that are used for verification. To address this problem, we investigate how to implement a secure face verification system that protects the facial images from being imitated. In our work, we use the DeepID2 convolutional neural network to extract the features of a facial image and an EM algorithm to solve the facial verification problem. To maintain the privacy of facial images, we apply homomorphic encryption schemes to encrypt the facial data and compute the EM algorithm in the ciphertext domain. We develop three face verification systems for surveillance (or entrance) control of a local community based on three levels of privacy concerns. The associated timing performances are presented to demonstrate their feasibility for practical implementation. © 2024, CC BY.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Remote Training in Task-Oriented Communication: Supervised or Self-Supervised with Fine-Tuning?

arXiv

引用

arXiv 2025年

作者： Li, Hongru Zhao, Hang He, Hengtao Song, Shenghui Zhang, Jun Letaief, Khaled B. Dept. of Electronic and Computer Engineering The Hong Kong University of Science and Technology Hong Kong

Task-oriented communication focuses on extracting and transmitting only the information relevant to specific tasks, effectively minimizing communication overhead. Most existing methods prioritize reducing this overhead during inference, often assuming feasible local training or minimal training communication resources. However, in real-world wireless systems with dynamic connection topologies, training models locally for each new connection is impractical, and task-specific information is often unavailable before establishing connections. Therefore, minimizing training overhead and enabling label-free, task-agnostic pre-training before the connection establishment are essential for effective task-oriented communication. In this paper, we tackle these challenges by employing a mutual information maximization approach grounded in self-supervised learning and information-theoretic analysis. We propose an efficient strategy that pre-trains the transmitter in a task-agnostic and label-free manner, followed by joint fine-tuning of both the transmitter and receiver in a task-specific, label-aware manner. Simulation results show that our proposed method reduces training communication overhead to about half that of full-supervised methods using the SGD optimizer, demonstrating significant improvements in training efficiency. Copyright © 2025, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Study of the contextual factors of a project affecting the build performance in Continuous Integration

Study of the contextual factors of a project affecting the b...

引用

IEEE International Conference on engineering Education: Innovative Practices and Future Trends (AICERA)

作者： Jetty Benjamin Juby Mathew Rubin Thottupurathu Jose Dept. of Computer Applications Amal Jyothi College of Engineering Kanjirappally Kerala India Research Scholar APJ Abdul Kalam Technological University Thiruvananthapuram Kerala Dept. of Computer Science and Engineering Amal Jyothi College of Engineering Kanjirappally Kerala India

Continuous Integration (CI) has a significant and substantial impact on software engineering. CI optimize the development processes, enhance the code quality and facilitate a collaborative environment resulting in the effective release of high quality software features to the end users. The performance and outcome of CI build is a major concern in CI environment. In this paper we address the impact of contextual factors of software projects that can influence the performance of build in Continuous Integration/Continuous Deployment (CI/CD) pipeline. We studied 588 large-scale GitHub projects that uses the Travis CI build system. An experimental study is conducted on these projects to leverage the history of prior build success and failures. Three ML models are trained on the data. The aim of the study was to identify the contextual factors of a project, influencing the performance of build. We used Logistic Regression, Decision Tree and Extreme Gradient Boost model to predict the build outcome. We identified that the contextual factors has a significant impact and the study will further helps to develop techniques to preemptively detect when a CI build environment is most likely to trigger a failure.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Lip Reading Into Text Using Deep Learning

Lip Reading Into Text Using Deep Learning

引用

International Conference on Confluence The Next Generation Information Technology Summit (Confluence)

作者： Krishna Manglani Utkarsh Bahuguna Subhash Chand Gupta dept. of Computer Science Amity School Of Engineering and Technology Uttar Pradesh India

An essential part of computer vision, lip reading, has grown significantly and is now used in autonomous driving, public safety, and hearing-impaired communication. This work provides an in-dept. analysis of both conventional and deep learning-based lip reading systems, examining their development over time, difficulties encountered, and critical role in human communication. The suggested LipNet model, which leverages the complementary abilities of Long Short-Term Memory networks (LSTMs) for temporal modelling and Convolutional Neural Networks (CNNs) for spatial feature extraction, is a central focus of this research. When LipNet is tested using the GRID Corpus dataset, it outperforms baseline models in terms of performance. Saliency maps and viseme analysis highlight LipNet's ability to capture phonologically relevant areas in lip motions, providing insight into the network's attention processes. Apart from examining the importance of datasets such as OuluVs2and GRID Corpus, this study presents GhostNet, a lightweight model that tackles the requirement for effective network designs. The testing results highlight LipNet's competitive Character Error Rate (CER) and Word Error Rate (WER) in comparison to baseline models, proving that it is the best method for real-world use. The research ends with a summary of potential future developments in lip reading technology, such as modal integration, realtime applications, personalisation, ethical issues, and the goal of creating a communication environment that is more inclusive. This study highlights the effectiveness of LipNet and paves the way for future developments in the ever-evolving area of lip reading by examining a variety of methodologies.

关键词：

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：