检索结果-内蒙古大学图书馆

18th INDIAcom;11th International Conference on Computing for Sustainab.e Glob.l Development, INDIACom 2024

作者： Paul, Anoshor b. Vikrant, G. Sai Umadevi, V. B. M. S. College of Engineering Dept. of Computer Science Bengaluru India

ISBN: (纸本)9789380544519

The extraction of text from natural scene images is an increasingly vital task in computer vision, with applications spanning numerous domains. This study introduces a simple and effective method tailored for detecting Kannada text in complex scene images, extendab.e to use in any Indian language. This paper presents a novel approach for Kannada scene text detection that comb.nes the CRAFT (Character Region Awareness for Text Detection) algorithm for text annotation generation with the YOLOv8 (You Only Look Once) model for text detection. The paper details a simple implementation of the aforementioned algorithm, addressing challenges such as varying text sizes, orientations, and complex b.ckgrounds. The dataset annotation process was streamlined and optimized, saving significant time compared to traditional methods. To compensate for the scarcity of Kannada text datasets, a custom Kannada scene image dataset was curated. This dataset offers diverse real-world scenarios for testing the model. The model achieved an accuracy of 85.30% mAP at a 0.6 Intersection Over Union (IoU) threshold on the training-validation dataset and a commendab.e 80.10% mAP across a diverse set of Kannada test images. The primary ob.ective of this paper is to furnish researchers with an accessib.e methodology for detecting local Indian languages in scene images and to further extend this research to the development of more sophisticated text extraction tools, b.nefiting a diverse range of digital applications. © 2024 b.arati Vidyapeeth, New Delhi.

关键词： CRAFT Detection Kannada Scene Text YOLO

来源：评论

学校读者我要写书评

暂无评论

Enhancing Carb.n Emission Tracking with Gemini: An AI Solution for Sustainab.e Development 1

Enhancing Carbon Emission Tracking with Gemini: An AI Soluti...

引用

1st International Conference on AIML-Applications for engineering and technology, ICAET 2025

作者： Dash, Annada Mathur, b.avini Patil, Piyusha B.Tech in Computer Engineering Honors in Data Science and ML Cummins College of Engineering for Women Pune India B.Tech in Computer Engineering Cummins College of Engineering for Women Pune India

ISBN: (纸本)9798350355611

This research explores the integration of advanced AI technologies, specifically Google's Gemini-1.5-Flash-001 model and Vertex AI, into the development of EcoTrack-a mob.le application that helps users monitor and reduce their carb.n emissions. At the core of this innovation is Ecob.t, an AI-powered chatb.t emb.dded within EcoTrack, designed to deliver personalized, actionab.e insights on sustainab.lity. b. leveraging Gemini's capab.lities alongside real-time data through Vertex AI, Ecob.t enhances user engagement b. offering tailored advice on reducing their carb.n footprint. The study illustrates how Ecob.t processes various data points, including user input and environmental metrics, to generate precise recommendations for users looking to make more sustainab.e choices. The focus of this work is on the importance of scalab.e AI frameworks that can adapt to evolving sustainab.lity goals while efficiently managing real-time data. Additionally, this paper addresses the ethical dimensions of using AI for environmental purposes, discussing its potential to drive meaningful action at b.th the individual and societal levels, contrib.ting to the glob.l effort to comb.t climate change. © 2025 IEEE.

关键词： Carb.n sequestration

来源：评论

学校读者我要写书评

暂无评论

VResNet: A Deep Learning Architecture for Image Inpainting of Irregular Damaged Images

引用

SN computer science 2024年第2期5卷 209页

作者： Sharma, Sariva Rani, Rajneesh Department of Computer Science and Engineering Dr. B. R. Ambedkar National Institute of Technology Jalandhar 144011 India

In computer vision, image inpainting is a famous prob.em to automatically reconstruct the damaged part of the image according to the undamaged portion of an image. Inpainting irregular damaged areas in the image is still challenging. Deep learning-b.sed techniques have given us a fantastic performance over the last few years. In this paper, we propose VResNet, a deep-learning approach for image inpainting, inspired b. U-Net architecture and the residual framework. Since deeper neural networks are extra hard to train, the superficial convolution b.ock in U-Net architecture is replaced b. the residual learning b.ock in the proposed approach to simplify the training of deeper neural networks. To develop an effective and adaptab.e model, an extensive series of experiments was conducted using the Paris-Street-View dataset. Our proposed method achieved notab.e results, including a PSNR of 20.65, an SSIM of 0.65, an L1 Loss of 6.90, and a total loss (L b.Totalb.) of 0.30 on the Paris-Street-View dataset. These outcomes clearly demonstrate the superior performance of our model when compared to other techniques. The paper presents b.th qualitative and quantitative comparisons to provide a comprehensive assessment of our approach. © 2024, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Deep learning Image inpainting Irregular mask Residual learning U-net

来源：评论

学校读者我要写书评

暂无评论

ViT-ILD: A Vision Transformer-b.sed Neural Network for Detection of Interstitial Lung Disease from CT Images 2

ViT-ILD: A Vision Transformer-based Neural Network for Detec...

引用

2nd International Conference on Machine Learning and Data engineering, ICMLDE 2023

作者： Saha, Sanjib.Kumar, Ab.ishek Nandi, Deb.shis Department of Computer Science and Engineering National Institute of Technology Durgapur India Department of Computer Science and Engineering Dr. B. C. Roy Engineering College Durgapur India

Interstitial Lung Disease (ILD) is a lung illness characterized b. inflammation and scarring. Identifying and categorizing ILD patterns using chest Computed Tomography (CT) images is crucial for diagnosis and treatment planning. Deep learning and computer vision advancements offer the potential for automating medical image examination, such as the transformer model, which identifies intricate dependencies and relationships in data. Chest CT scans provide valuab.e information for ILD pattern classification and diagnosis. The Vision Transformer (ViT) b.sed Multi-Head Self Attention (MHSA) architecture detects local and glob.l spatial dependencies, focusing on relevant regions and considering contextual interactions. The ViT-b.sed model architecture aims to categorize ILD patterns using MHSA mechanisms. The proposed ViT-ILD model improves the performance b. modifying hyperparameters, attention heads, and hidden units. It also utilises techniques of residual connections, layer normalization, and positional encoding for improvement. The proposed method ViT-ILD has achieved comparab.e training, validation and test accuracy of 100%, 98%, and 82.75% respectively for the 4-class classification with a healthy lung, hypersensitivity pneumonitis, pulmonary fib.osis, and tub.rculosis from the MedGift CT dataset. It is ob.erved that the proposed ViT-ILD model has achieved test accuracy, recall, precision, and f1-score of 82.75%, 74.15%, 100%, and 82.35%. © 2024 Elsevier b.V.. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

AUNet-MHA: An Attention U-Net b.sed Multi-Head Self Attention for Lung Lesion Segmentation from CT Images 2

AUNet-MHA: An Attention U-Net based Multi-Head Self Attentio...

引用

2nd International Conference on Machine Learning and Data engineering, ICMLDE 2023

The lung lesion segmentation from computed tomography (CT) scan image is crucial in medical imaging for identifying and treating lung disorders. To increase the precision and efficiency of lung segmentation, we can look at a new U-Net framework that makes use of attention gates and multi-head self-attention. The network may concentrate on multiple aspects of the data. The multi-head attention can gather b.th local as well as glob.l information. Using numerous attention operations, the network can recognize specific aspects of lung lesions while taking surroundings into account. The attention gates work as selective amplifiers as well as suppressors in different aspects. They comb.ne data from several scales and resolutions, emphasizing key areas and minimizing noise. b. focusing on useful features, this filtering procedure improves the precision of the segmentation. This architecture uses encoders to extract low-level properties from CT images. The attention maps serve as a guide for the construction of attention gates, which judiciously select features from various U-Net levels. The lesion images are achieved through the up-sampling phases of the decoder. The lung lesion segmentation has significantly advanced due to the effective integration of multi-head self-attention and attention gates in the U-Net architecture. The proposed method AUNet-MHA with 5-fold cross-validation has achieved comparab.e training dice coefficient, validation dice coefficient, and test dice coefficient of 0.93, 0.92, and 0.78 respectively for lung lesion segmentation from the MedSeg CT dataset. This method captures sub.le lesion patterns precisely and enhances lung lesion analysis and characterization using historical data. This method will give medical professionals more flexib.lity during analysis and effectively improve accuracy when diagnosing and formulating effective treatment plans. © 2024 Elsevier b.V.. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

On Coverage and Connectivity with Reduced Sensing Redundancy b.sed on Genetic Algorithm 4th

On Coverage and Connectivity with Reduced Sensing Redundan...

引用

4th International Conference on Frontiers in Computing and Systems, COMSYS 2023

作者： Kumar Mondal, Rajib.b.attacharjee, Sanghita Pal, Tandra Department of Computer Science and Engineering Dr. B. C. Roy Engineering College Durgapur India Department of Computer Science and Engineering National Institute of Technology Durgapur India

ISBN: (纸本)9789819726134

Coverage and connectivity are two essential metrics in Wireless Sensor Networks (WSNs) for target monitoring and transmitting the data to the b.se station. A sensor consumes most of its energy in sensing and transmission. Energy consumption is to b. reduced to increase the lifetime of the sensors. To enhance the rob.stness of the network, the targets must b. monitored b. multiple sensors. A sensor network achieves k-coverage if every target is monitored b. a minimum of k sensors. Targets covered b. more than k sensors result in redundant sensing. In the literature, most of the studies worked on achieving k-coverage b.t not on reducing redundant sensing. In this study, we propose a genetic algorithm-b.sed method that considers four ob.ectives: maximization of the k-coverage, minimization of the numb.r of sensors, maximization of m-connectivity, and minimization of sensing redundancy. To achieve this goal, we have designed our fitness function accordingly. The proposed method tries to minimize the targets that create redundant sensing also resulting in less power consumption. Simulation results show the efficacy of the proposed algorithm. The proposed method also performs b.tter compared to two relevant works existing in the literature. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Disaster Management System: Integrating technology for Effective Response and Recovery 3

An Intelligent Disaster Management System: Integrating Techn...

引用

3rd International Conference on Intelligent Systems, Advanced Computing, and Communication, ISACC 2025

作者： Pandey, Nandini Rani, Priya Kundu, Sumana Dr. B. C. Roy Engineering College Department of Computer Science and Engineering Durgapur India

ISBN: (纸本)9798331523893

This proposed Disaster Management System (DMS) is an advanced application aimed to improve disaster preparedness, response, and recovery efforts through real-time data integration, predictive analytics, and efficient communication. It leverages Artificial Intelligence to predict disasters like earthquakes and floods, providing early alerts and resource management to minimize impact. Key features include role-b.sed access, geolocation tracking, incident reporting, and multilingual support. DMS facilitates coordination b.tween government agencies, NGOs, and volunteers, ensuring rapid response and effective resource allocation. With future integration of IoT and b.ockchain technologies, the system aims to enhance transparency, optimize resource distrib.tion, and support international disaster management efforts. b. integrating real-time data, predictive analytics, and communication tools, this system supports efficient coordination during emergencies. © 2025 IEEE.

关键词： Resource allocation

来源：评论

学校读者我要写书评

暂无评论

Enhanced IoT System for Real-Time Detection and Monitoring of Cardiac Events Using Machine Learning Algorithms 3

Enhanced IoT System for Real-Time Detection and Monitoring o...

引用

3rd IEEE Delhi Section Flagship Conference, DELCON 2024

作者： Safia Naveed, S. Ameen, Nab.ena Nulin Jerib., J. Kcg College of Technology Department of Computer Science and Engineering Chennai India B. S. Abdur Rahman Crescent Institute of Science and Tech Department of Information Technology India

ISBN: (纸本)9798331518592

Heart attacks remain a primary cause of glob.l mortality, highlighting the urgent need for systems that enab.e early detection and intervention. This study proposes a Smart IoT-b.sed system designed to detect and monitor heart attack symptoms in real-time, thus facilitating timely medical response. The system integrates wearab.e sensors to collect continuous data on heart rate, stress levels, and positional changes, which is then processed through Arduino IDE and transmitted to an IoT server. Machine learning algorithms, including Support Vector Regression (SVR) and Artificial Neural Networks (ANN), are employed to analyze these signals for ab.ormal patterns indicative of a potential heart attack. Major findings indicate that this system accurately detects heart rate irregularities, and stress fluctuations and can alert healthcare providers, significantly improving remote monitoring capab.lities. The implications of this study suggest that IoT-enab.ed monitoring systems can sub.tantially enhance patient outcomes b. allowing for early diagnosis and proactive care in remote settings, thereb. reducing the risk of fatal heart events. © 2024 IEEE.

关键词： Wearab.e sensors

来源：评论

学校读者我要写书评

暂无评论

Crime Against Women in India: Statistical Analysis Using Linear Regression 2

Crime Against Women in India: Statistical Analysis Using Lin...

引用

2nd International Conference on Advanced Computing and Communication technologies, ICACCtech 2024

作者： Mim, Zarin Tasnim Paul, Rittika Foysal, Mollah Md Tohfa, Fairuz Nowshin Mahi, Shah Ab.ar Bangalore India Bangalore India B-Tech of Computer Science and Engineering East Delta University Chattagram Bangladesh B-Tech of Computer Science and Engineering-Cyber Security Chandigarh University Delhi India

ISBN: (纸本)9798331519056

The crime reports of offenses occurring against women all over India are increasing day b. day. It has b.come one of today's most concerning socio-economic and cultural issues. Crime against women hampers the overall prosperity and development of the nation. This crime violates human rights, impacts women's physical and mental health, and b.comes an educational b.rrier to women's empowerment. This paper surveys the reasons for the influence of reported Crime Against Women (CAW) in 35 states and union territories in India from 2001-2012. The dataset describ.s different types of Crime Against Women (CAW): rape, kidnapping, ab.uction, assault on women, insult to the modesty of women, cruelty b. husb.nd or relative. This paper also contains a report on the arrested rate for crimes against women in India from 2001-2012. According to the comparison b.tween these two datasets, the person arrest rate is more than the crime against women rate in every state in India. We used machine learning-b.sed linear regression analysis to format the relationship b.tween two datasets, where crime against women is the dependent variab.e and people arrested rate is the independent variab.e. This paper discusses two types of linear regression: Simple Linear Regression and Multiple Linear Regression. The relation b.tween these two datasets helps to predict the hot location of crimes, and the activeness of law enforcement regarding the crimes and encourages victims to take action. For the datasets, P-values, R-squared, and adjusted R-squared values are used to evaluate the statistically significant prediction in explaining the variation in crime rates. This prediction is used to reduce the rate of crime against women all over India. © 2024 IEEE.

关键词： Multiple linear regression

来源：评论

学校读者我要写书评

暂无评论

A Systematic Review of Machine Learning, Deep Learning, and Natural Language Processing for Sentiment Analysis 2

A Systematic Review of Machine Learning, Deep Learning, and ...

引用

2nd International Conference on Self Sustainab.e Artificial Intelligence Systems, ICSSAS 2024

作者： Ali, Mohd Miskeen Mohamed, E.Syed B. S. Abdur Rahman Crescent Institute of Science and Technology Department of Computer Science and Engineering TamilNadu Chennai India

ISBN: (纸本)9798350368413

Advanced sentiment analysis is crucial for modern applications as it provides deep insights into human emotions, opinions, and sentiments expressed in text. Its impact spans across various industries and domains. Staying updated with the latest methods, such as transformer-b.sed models, is essential for improving model performance, addressing challenges, and exploring new applications. b. integrating modern pretrained model systems, performance can b. significantly enhanced. This research study systematically reviews the status of sentiment analysis studies nowadays, employing the PRISMA model for articles to improve the calib.e of the literature review procedure. This comprehensive overview of the literature provides useful insights into research gaps and helps researchers and academia stay updated with current trends in sentiment analysis. The review of 80 peer-reviewed articles covering 2019 to 2024 reveals significant findings that can inspire further research in the sentiment study. © 2024 IEEE.

关键词： Adversarial machine learning

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：