检索结果-内蒙古大学图书馆

Journal of the Faculty of engineering and Architecture of Gazi University 2023年第4期38卷 2579-2593页

作者： Dölek, İshak Kurt, Atakan Mina R&D Informatics Entertech Office:322 Istanbul- University Cerrahpaşa Istanbul34320 Turkey Department of Computer Engineering Faculty of Engineering Istanbul University Cerrahpaşa Istanbul34320 Turkey

In this paper, we present a web-based optical character recognition (OCR) system that converts images of Ottoman documents printed with naskh font into text using CNN+RNN-based deep neural network models. For training, three datasets - original, synthetic, and hybrid - were prepared and three different OCR models were created. The original data set consists of 1,000 pages and the synthetic data set consists of 23,000 pages. Hybrid data set contains both. The trained models were compared with Tesseract's Arabic and Persian, Google Docs' Arabic, Abby FineReader's Arabic, and Miletos OCR model/tools with a 21-page test set. The comparison was made with 3 different texts (raw, normalized, and joined) and using 3 different criteria (character, ligature, and word recognition). The *** Hybrid model produced significantly better results than the others with 88.86% raw, 96.12% normalized, and 97.37% joined accuracy in character recognition;80.48% raw, 91.60% normalized, and 97.37% joined accuracy in ligature recognition;and 44.08% raw and 66.45% normalized accuracy in word recognition. To investigate the effects of the characteristics of the alphabet on OCR, character, ligature, and word frequency analyses of Ottoman was performed. In this analysis, the characters in the alphabet were grouped according to distinctive features such as connectedness, letter body, position and number of dots, type of character, and source language;and frequencies and recognition accuracies were examined for each group. OCR results are also reported for each character. © 2023 Gazi Universitesi Muhendislik-Mimarlik. 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

Preface for the 10th International conference on sociotechnical perspectives in Information Systems (STPIS 2024)

引用

CEUR Workshop Proceedings 2024年 3857卷 1-7页

作者： Kävrestad, Joakim Bednar, Peter Rajanen, Mikko School of Engineering Jönköping University Sweden Department of Informatics University of Lund Sweden School of Computing University of Portsmouth United Kingdom INTERACT Research Unit University of Oulu Finland

来源：评论

学校读者我要写书评

暂无评论

Developing Sustainable Solutions Using Technological Approaches for Disaster Management and Energy Access in Mountainous Ranges: A Case Study of Sera Village, India 4

Developing Sustainable Solutions Using Technological Approac...

引用

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

作者： Kumar, Y. Jagadish Surya Vamsi, D. Nishaanth, K.N. Mouria, Naren Raghul, K.R. Sampath Kumar, S. Cornet, Souresh Coimbatore Amrita Vishwa Vidyapeetham Amrita School of Computing Department of Computer Science and Engineering India Coimbatore Amrita Vishwa Vidyapeetham Amrita School of Engineering Department of Electrical and Computer Engineering India Coimbatore Amrita Vishwa Vidyapeetham Amrita School of Engineering Department of Electronics and Communication Engineering India Coimbatore Amrita Vishwa Vidyapeetham Amrita School of Engineering Department of Electrical and Electronics Engineering India Amrita Vishwa Vidyapeetham Amritapuri Amrita School for Sustainable Futures India

ISBN: (纸本)9798350343670

This research looks at two essential aspects of sustainable development in India's mountainous regions: energy access and vulnerability to natural disasters such as landslides. The research is based on an inductive in-depth evaluation of Sera village in Uttarakhand using Participatory Rural Appraisal (PRA) tools and Human-Centered Design (HCD) techniques. This approach enables a thorough understanding of these regions' complex challenges in order to provide relevant technological alternatives to meet the identified challenges. An integrated solution for landslide prediction powered by a hydropower system is presented, incorporating image processing and deep learning. This research provides useful insights for future work to address the issues inhabitants face of mountainous settings, ultimately supporting sustainable development in rural areas. © 2024 IEEE.

关键词： Deep learning Image Processing Landslides Live-in-Labs® Participatory Rural Appraisal Rural Electrification

来源：评论

学校读者我要写书评

暂无评论

Face Mask detection using Mask R-CNN to control the spread of Covid-19

Face Mask detection using Mask R-CNN to control the spread o...

引用

2023 IEEE International Conference on Recent Trends in Electronics and Communication, ICRTEC 2023

作者： Mohan, V.S. Abhay Gowda, G. Nair, Rekha R Kishore, S. Babu, Tina School of Engineering Dayananda Sagar University Department of Computer Applications Banglore India School of Engineering Dayananda Sagar University Department of Computer Science and Engineering Banglore India

ISBN: (纸本)9798350396195

Business executives are developing cutting-edge digital solutions as the virus outbreak spreads. A face mask detection system is one of them, and it can be used to spot people wearing them. Face mask identification software and applications have already been released by a few businesses, and others have promised to do the same for the service. The proposed work examines face mask detection accuracy using CNN networks. Mask wear is now required in many developed and developing countries worldwide when leaving the house or entering public spaces. It will be difficult to maintain touchless access control in buildings while recognising faces wearing masks on any surveillance systems. Masks covering faces has made face detection algorithms and performance difficult. The proposed work detect face mask labeled no mask or mask with detection accuracy. The work train the system to click images of a face and provide labeled data. The work is classified using Convolution Neural Network (CNN), a Deep learning technique, to classify the input image with the help of the classification algorithm MobileNetV2. The trained system shows whether a person in the video frame is wearing a mask or not. © 2023 IEEE.

关键词： Access control

来源：评论

学校读者我要写书评

暂无评论

Performance Analysis of Machine Learning Models for Liver Disease Patient Classification

Performance Analysis of Machine Learning Models for Liver Di...

引用

2024 International Conference on Intelligent Computing and Sustainable Innovations in Technology, IC-SIT 2024

作者： Behera, Ranjit Kumar Behera, Dayal Kumar Swetanisha, Subhra Jena, Ajay Kumar Silicon University Department of Computer Science & Engineering Bhubaneswar India KIIT Deemed to Be University School of Computer Engineering Bhubaneswar India Trident Academy of Technology Department of CSE Odisha Bhubaneswar India KIIT University School of Computer Engineering Odisha Bhubaneswar India

ISBN: (纸本)9798350369175

Liver disease is a significant global health concern, necessitating the development of accurate and efficient diagnostic tools. This study presents a comparative analysis of various machine learning classifiers, including Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), XGBoost, and Ensemble Models, for the classification of liver disease patients. The challenge of class imbalance in the dataset is addressed using Synthetic Minority Over-sampling Technique (SMOTE), ensuring a more balanced and representative training process. Missing data is handled with the K-Nearest Neighbors (KNN) Imputer, while Recursive Feature Elimination (RFE) is employed to select the most relevant features, optimizing the model's performance. Experimental results demonstrate the effectiveness of these approaches, with a detailed comparison of each classifier's performance in terms of accuracy and F1-score. This study provides valuable insights into the strengths and limitations of different machine learning classifiers for liver disease classification, contributing to the advancement of predictive modeling in healthcare. © 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

Enhancing Multimodal Rumor Detection with Statistical Image Features and Modal Alignment via Contrastive Learning 21st

Enhancing Multimodal Rumor Detection with Statistical Image...

引用

21st Pacific Rim International Conference on Artificial Intelligence, PRICAI 2024

作者： Zhou, Chenyu Li, Xiuhong Li, Zhe Chen, Fan Sheng, Jiabao Chen, Bin Wang, Haoyu School of Computer Science and Technology Xinjiang University Xinjiang China Department of Electrical and Electronic Engineering The Hong Kong Polytechnic University Hong Kong Department of Electrical Engineering Stanford University CA United States Department of Health Technology and Informatics The Hong Kong Polytechnic University Hong Kong School of Mechanical Engineering University of Shanghai for Science and Technology Shanghai China School of Electronic and Electrical Engineering Shanghai University of Engineering Science Shanghai China

ISBN: (纸本)9789819601219

The swift proliferation of multimodal rumors on social media, particularly those with manipulated images and complex intermodal interactions, significantly challenges current detection methods. In response, we utilize statistical image features, including mean and variance, to capture spatial attributes effectively and improve the detection of image-tampered tweets. To tackle complex intermodal correlations, we introduce a contrastive learning approach that aligns features across modalities efficiently. Additionally, we introduce a cross-attention fusion module (CAFM) that enhances the integration of image and text modalities, thereby improving multimodal rumor detection performance. In conclusion, we propose the cross-attention fusion network (ConCAFN), leveraging contrastive learning for robust multimodal rumor detection. Extensive experiments on two real-world datasets confirm the model’s enhanced capability to detect multimodal rumors accurately, demonstrating our methods’ effectiveness. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Spatial Attention Integrated EfficientNet Architecture for Breast Cancer Classification with Explainable AI

引用

computers, Materials & Continua 2024年第9期80卷 5029-5045页

作者： Sannasi Chakravarthy Bharanidharan Nagarajan Surbhi Bhatia Khan Vinoth Kumar Venkatesan Mahesh Thyluru Ramakrishna Ahlam AlMusharraf Khursheed Aurungzeb Department of Electronics and Communication Engineering Bannari Amman Institute of TechnologySathyamangalam638402India School of Computer Science Engineering and Information Systems(SCORE) Vellore Institute of TechnologyVellore632014India School of Science Engineering and EnvironmentUniversity of SalfordManchesterM54WTUK Department of Computer Science&Engineering Faculty of Engineering and TechnologyJAIN(Deemed-to-be University)Bengaluru562112India Department of Management College of Business AdministrationPrincess Nourah Bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityP.O.Box 51178Riyadh11543Saudi Arabia Adjunct Research Faculty Centre for Research Impact&OutcomeChitkara UniversityRajpura140401India

Breast cancer is a type of cancer responsible for higher mortality rates among *** cruelty of breast cancer always requires a promising approach for its earlier *** light of this,the proposed research leverages the representation ability of pretrained EfficientNet-B0 model and the classification ability of the XGBoost model for the binary classification of breast *** addition,the above transfer learning model is modified in such a way that it will focus more on tumor cells in the input ***,the work proposed an EfficientNet-B0 having a Spatial Attention Layer with XGBoost(ESA-XGBNet)for binary classification of *** this,the work is trained,tested,and validated using original and augmented mammogram images of three public datasets namely CBIS-DDSM,INbreast,and MIAS *** accuracy of 97.585%(CBISDDSM),98.255%(INbreast),and 98.91%(MIAS)is obtained using the proposed ESA-XGBNet architecture as compared with the existing ***,the decision-making of the proposed ESA-XGBNet architecture is visualized and validated using the Attention Guided GradCAM-based Explainable AI technique.

关键词： EfficientNet mammograms breast cancer Explainable AI deep-learning transfer learning

来源：评论

学校读者我要写书评

暂无评论

A comprehensive review of effective parameters to improve the performance of the Savonius turbine using a computational model and comparison with practical results

引用

Water-Energy Nexus 2024年第1期7卷 266-274页

作者： Jebelli, Ali Lotfi, Nafiseh Saeid Zare, Mohammad Yagoub, Mustapha C.E. Department of Engineering and Design Robotic Century Inc Alberta Canada Faculty of New Science and Technologies Tehran University Tehran Iran School of Electrical Engineering and Computer Science University of Ottawa ON Canada

Amidst growing global concerns over climate change and escalating greenhouse gas emissions from fossil fuels, the pursuit of renewable energy sources has become critical. This study focuses on harnessing hydropower using Savonius turbines, which are known for their efficiency in generating energy at lower flow rates. However, the intrinsic low efficiency of these turbines necessitates precise optimization tailored to specific river or channel conditions. In this research, we optimized the performance of Savonius turbines by analyzing key parameters such as the height-to-diameter ratio, blade twist, and the integration of multi-stage configurations with deflectors. Our findings reveal significant efficiency improvements through strategic modifications. Specifically, by reducing the height-to-diameter ratio from 1.2 to 0.8 and maintaining a Tip Speed Ratio (TSR) of 0.6, the power coefficient increased by 15%, from 0.33 to 0.38. Further optimization was achieved by adjusting the blade twist angle, with an increase in power coefficient up to an optimal angle of 45°, beyond which efficiency declined. Implementing a two-stage turbine setup with a 90-degree phase difference between stages further improved the power coefficient to 0.51 at the same TSR. Additionally, the use of deflectors, particularly at a 90° angle, significantly boosted the power coefficient, highlighting their effectiveness in optimizing water flow impact on the turbine. This comprehensive study not only advances the understanding of Savonius turbine optimization but also contributes to broader renewable energy applications. The research offers critical insights into sustainable hydroelectric power generation, providing practical solutions to enhance turbine performance for real-world applications. © 2024

关键词： Clean energy

来源：评论

学校读者我要写书评

暂无评论

Interpretation of Sign Language Using Machine Learning 2

Interpretation of Sign Language Using Machine Learning

引用

2nd International Conference Computational and Characterization Techniques in engineering and Sciences, IC3TES 2024

作者： Srivastava, Palak Hazela, Bramah Singh, Shikha Asthana, Pallavi Pandey, Deependra Singh, Kamlesh Kumar Singh, Vineet Amity University Amity School of Engineering & Technology Department of Computer Science Engineering Uttar Pradesh India Amity University Amity School of Engineering & Technology Department of Electronics & Communication Engineering Uttar Pradesh India

ISBN: (纸本)9798350364699

Sign language detection using machine learning has emerged as a crucial area of research aimed at bridging communication barriers between individuals with hearing impairments and the broader community. This paper explores the application of machine learning techniques for the automatic recognition and interpretation of sign language gestures. A systematic review of existing methodologies, including traditional machine learning approaches and recent advancements in deep learning architectures, reveals key challenges such as variability in signing styles, regional dialects, and real-time processing requirements. To address these challenges, we propose a novel approach that leverages convolutional neural networks (CNNs) and long short-term memory (LSTM) models, which have demonstrated significant improvements in accuracy and generalization across diverse datasets. Our experimental results show the efficacy of these models on benchmark sign language datasets, achieving high levels of accuracy and robustness in real-world conditions. Furthermore, this study discusses the potential societal impacts of sign language detection systems in education, healthcare, and everyday communication, emphasizing the role of technological innovation in fostering inclusivity. By advancing machine learning methodologies, we contribute to the development of accessible communication solutions for individuals with hearing impairments. © 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

Network Anomaly Detection using Autoencoder on Various Datasets: A Comprehensive Review

引用

Recent Patents on engineering 2024年第9期18卷 63-77页

作者： Singh, Richa Srivastava, Nidhi Kumar, Ashwani Amity Institute of Information Technology Amity University Lucknow Campus Uttar Pradesh Lucknow India Department of Computer Science and Engineering School of Engineering and Technology Sharda University Greater Noida India

The scientific community is currently very concerned about information and communication technology security because any assault or network anomaly can have a remarkable collision on a number of areas, including national security, the storage of private data, social welfare, economic concerns, and more. As a result, many strategies and approaches for this goal have been developed over time, making the anomaly detection domain a large research subject. The primary concern of this patent study is to review the most crucial elements relating to anomaly detection, including an overview of background analysis and a core study on the most important approaches, procedures, and systems in the field. To make the structure of this survey easier to understand, the domain of anomaly detection was examined along with five dimensions: Detection methods in network traffic, objectives of the patent paper, various datasets used, accuracy, and open issues/gaps. The gap which has been identified after the survey can be extended as a future scope might be helpful for the researcher. © 2024 Bentham Science Publishers.

关键词： Anomaly detection

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：