检索结果-内蒙古大学图书馆

A Text Detection Method Based on Multi-Scale Selective Fusion Feature Pyramid and Multi-Semantic Spatial Network for Visual IoT

引用

IEEE Internet of Things Journal 2025年第11期12卷 15997-16008页

作者： Wang, Manli Dou, Zeya Chen, Wei Wang, Zehua Yu, Fei Richard Leung, Victor C. M. Henan Polytechnic University School of Physics and Electronic Information Henan Jiaozuo454003 China China University of Mining and Technology School of Computer Science and Technology Jiangsu Xuzhou221116 China China University of Mining and Technology School of Mechanical Electronic and Information Engineering Beijing100083 China The University of British Columbia Department of Electrical and Computer Engineering VancouverBCV6T 1Z4 Canada Carleton University School of Information Technology OttawaONK1S 5B6 Canada

With the rapid development of Visual Internet of Things (VIoT) and text detection technology, they have been widely combined and applied to many industrial production sites, such as label text detection, achieving impressive results. However, there are still many shortcomings in the text detection technology: 1) The existing VIoT system has very limited detection precision for text with large scale changes, especially for some small-scale text detection;2) The existing text detection algorithms cannot meet the actual situation, as the labels often contain handwritten texts, and the text to be detected is arbitrary shape;3) In the actual detection, there are many creases or defects on the text label. To solve the above problems, this paper designs a text detection method based on a multi-scale selection fusion feature pyramid and multi-semantic spatial network to assist the VIoT system in detecting label text. Firstly, a multi-scale selective fusion feature pyramid is designed, which not only uses the texture extraction module to effectively improve the text texture feature and multi-scale feature extraction ability, but also uses the cross-scale selective fusion block to selectively fuse the features of different stages to reduce the influence of pollution on detection. In addition, a multi-semantic spatial network is designed to capture the multi-semantic spatial information of each feature channel by using the multi-scale deep shared one-dimensional convolution, which effectively integrates global context dependence and multi-semantic spatial prior. Experimental results show that the comprehensive index F-measure on the public datasets ICDAR2015, Total-Text and CTW1500 is increased by 5.7%, 3.3%and 3.8% respectively. Furthermore, the precision, recall, and F-measure on the dataset Label-Text are 94.6%, 90.7%and 92.6% respectively. The label text detection VIoT system we designed has been deployed in the field and achieved excellent performance. The code of our pro

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Open Source Sounding Rocket-based Free-Falling Platform to Conduct Reduced Gravity Research

Open Source Sounding Rocket-based Free-Falling Platform to C...

引用

2024 IAF Microgravity Sciences and Processes Symposium at the 75th International Astronautical Congress, IAC 2024

作者： Dalla Pria, G.L. Brückner, V. Åkerlund, B. Almqvist, E. Arancibia Parra, L. Broström, A. Buck, C. Gels, J. Hollmark, A. Höglund, F. Jakobsson, A. Korotkijs, A. Malmqvist, A. Mårtensson, J. Kuhn, T. Department of Computer Science Electrical and Space Engineering Luleå University of Technology Bengt Hultqvists väg 1 Kiruna981 92 Sweden

ISBN: (纸本)9798331312077

Free fall Experiment uNit for Reduced gravity Investigations and Research (FENRIR) is a Rocket Experiment for University Students (REXUS) project, based at the Kiruna Space Campus of Lulea University of technology (LTU), which will fly onboard a sounding rocket in March 2025. It offers an easy-to-use platform for future free-falling experiments that need to be ejected from sounding rockets. A Free Falling Unit (FFU) will be ejected from the rocket just before its apogee at approximately 70 km, ensuring around 70 seconds of reduced gravity. To increase the quality of reduced gravity conditions, an Attitude Stabilisation System consisting of three reaction wheels is included in the FFU. To prove its feasibility, the FFU hosts an experiment related to Phase Change Materials during its maiden flight. The behaviour of the substance is expected to vary due to the absence of buoyancy convection in reduced gravity conditions. This scientific experiment will serve as an example of the many possible scientific investigations that can be performed on the platform. After experiencing reduced gravity, the FFU recovery system will deploy a parachute and relay location data to the users to ensure safe recovery of the experiments inside. Finally, after FENRlR's maiden flight and post-flight activities, the entire project will be published as open source and the physical structure will be inherited by LTU to be potentially reused. Upper secondary and university students are foreseen as potential end-users: thanks to FENRlR's apparatus, they will have the opportunity to focus only on the mission's experiment and accompanying electronics, without having to design the entire structure from scratch. The platform is designed modularly to ensure ease of use and adjustability for different needs and use cases. This paper summarises the development of the FENRIR platform and how internal and external stakeholders and partners have driven the design and implementation process. Copyright © 20

关键词： Sounding rockets

来源：评论

学校读者我要写书评

暂无评论

Advanced Real-Time Patient Health Monitoring System 26

Advanced Real-Time Patient Health Monitoring System

引用

26th International Conference on computer and Information technology, ICCIT 2023

作者： Islam, Saiful Billah, Mohtasim Biswas, Richard Victor Das, Santu Faculty of Science & Technology Department of Computer Science & Engineering 408/1 Kuratoli Dhaka1229 Bangladesh Faculty of Engineering Department of Electrical & Electronic Engineering 408/1 Kuratoli Dhaka1229 Bangladesh

ISBN: (纸本)9798350359015

The rapid advancement of technology has brought transformative changes in the field of the healthcare system in recent years. Since they can enhance the quality and accessibility of healthcare services, particularly for patients with chronic conditions or in rural regions, microcontroller-based health monitoring systems have been extensively researched and developed in recent years. This paper presents a system capable of multiple measurements and improved ability of health parameters with a cost-efficient solution with reliability and portability. The proposed system is able to measure the parameters at an accuracy of 97.8% in terms of heart rate, blood pressure, body temperature, and oxygen saturation. The system also proposes continuous monitoring of the patient's health status by collecting data from sensors attached to the patient's finger and transmitting it wirelessly to a central database via Blynk where the healthcare provider can monitor the results in real-time and provide immediate feedback. © 2023 IEEE.

关键词： Arduino Mega 2560 R3 Blood pressure Blynk ESP8266 Heart rate MAX30102 SpO2 WLAN

来源：评论

学校读者我要写书评

暂无评论

AutoML Trading: A Rule-Based Model to Predict the Bull and Bearish Market

引用

Journal of The Institution of Engineers (India): Series B 2024年第4期105卷 913-928页

作者： Padhi, Dushmanta Kumar Padhy, Neelamadhab Panda, Baidyanath Bhoi, Akash Kumar School of Engineering and Technology Department of Computer Science and Engineering GIET University Gunupur India LTIMindtree 1 American Row 3rd Floor HartfordCT06103 United States Directorate of Research Sikkim Manipal University Sikkim Gangtok737102 India

Many researchers from different fields have tried to figure out how to forecast the future and trade on the share market since it is hard to accomplish because the market is so complicated. Machine-learning techniques have been exceptionally productive in this domain in recent times. Financial experts are creating a "trading system" for stock returns and trend forecasting. Since "trading online" was introduced, the amount of transactions that take place on the equity market each day has seen a significant increase. The objective of this study is to propose a rule-based trading framework not only forecasts the market but also finds out the best bull and bearish weekdays for trading. So here in this framework, we proposed an automated machine-learning (AutoML) online trading system that combines two modules. The first module finds the best weekdays to be bull or bearish with an algorithmic trading strategy. The second module combines two sub-models: an ensemble-based classification module and an ensemble-based regression module. The bagging classifier with the Extra Tree classifier is used as the classification model, whereas the bagging regressor with SGD regressor is used for the regression model. Different matrices were followed as per the criteria with Nifty-50 indices for model performance evaluation purposes by us. The regression model has an average RMSE value of 0.2091 MSE value of 0.0551, and the classification model’s accuracy range varies between 95.65 and 99.59. With the help of this framework, even if a little knowledge, investors can minimize investment loss and maximize their profit. © The Institution of Engineers (India) 2024.

关键词： Financial markets

来源：评论

学校读者我要写书评

暂无评论

Quantum-Assisted activation for supervised learning in healthcare-based intrusion detection systems

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第3期5卷 977-984页

作者： Laxminarayana, Nikhil Mishra, Nimish Tiwari, Prayag Garg, Sahil Behera, Bikash K. Farouk, Ahmed Department of Electronics and Communications Engineering Indian Institute of Information Technology West Bengal 741235 India Department of Computer Science and Engineering Indian Institute of Information Technology West Bengal 741235 India Department of Computer Science Aalto University Espoo02150 Finland Ecole de Technologie Superieure MontrealQCH3C 1K3 Canada Pvt. Ltd. WB Mohanpur741246 India Department of Computer Science Faculty of Computers and Artificial Intelligence South Valley University Hurghada83523 Egypt

Intrusion detection systems (IDSs) are amongst the most important automated defense mechanisms in modern industry. It is guarding against many attack vectors, especially in healthcare, where sensitive information (patients medical history, prescriptions, electronic health records, medical bills/debts, and many other sensitive data points) is open to compromise from adversaries. In the big data era, classical machine learning has been applied to train IDS. However, classical IDS tend to be complex: either using several hidden layers susceptible to overfitting on training data or using overly complex architectures such as convolutional neural networks, long-short term memory systems, and recurrent neural networks. This article explored the combination of principles of quantum mechanics and neural networks to train IDS. A hybrid classical-quantum neural architecture is proposed with a quantum-Assisted activation function that successfully captures patterns in the dataset while having less architectural memory footprint than classical solutions. The experimental results are demonstrated on the popular KDD99 dataset while comparing our solution to other classical models. Impact Statement Intrusion detection systems are dynamic defenses against network breach attacks. Lately, machine learning has been leveraged to perform automated intrusion detection. However, classical machine learning needs a large amount of data and overly complex architectures to "learn" patterns from data. In this work, we leverage machine learning concepts to derive a novel quantum activation function that greatly simplifies neural architectural complexity while achieving same level of accuracy and performance. Our architecture is much simpler than state-ofthe-Art based classical systems, simpler to train, and easy to handle. © 2024 Institute of Electrical and Electronics Engineers Inc.. All rights reserved.

关键词： Network architecture

来源：评论

学校读者我要写书评

暂无评论

Exploring Multimodal Framework of Optimized Feature-Based Machine Learning to Revolutionize the Diagnosis of Parkinson’s Disease: AI-Driven Insights

引用

Biomedical Materials and Devices 2025年 1-20页

作者： Khanom, Fahmida Mostafiz, Rafid Uddin, Khandaker Mohammad Mohi Department of Mathematics American International University-Bangladesh 408/1 Kuratoli Khilkhet Dhaka-1229 Dhaka Bangladesh Institute of Information Technology Noakhali Science and Technology University Noakhali Bangladesh Department of Computer Science and Engineering Southeast University Dhaka China

Parkinson’s disease (PD), a progressive neurodegenerative disorder, poses significant challenges to healthcare systems worldwide due to its increasing prevalence and need for early detection to improve patient outcomes. This study introduces a robust diagnostic framework leveraging a multimodal feature selection approach, combining Minimum Redundancy Maximum Relevance (mRMR) with Recursive Feature Elimination (RFE), to optimize predictive accuracy in detecting PD. Using a dataset of 894 instances from the UCI Machine Learning Repository, the framework implemented comprehensive data preprocessing and feature selection techniques to refine input features. The Extra Trees classifier, integrated with the mRMR-RFE multimodal feature selection method, demonstrated superior performance over nine other machine learning models, achieving a perfect accuracy of 100%, with high specificity, sensitivity, precision, and F1 scores. This framework highlights the efficacy of ensemble learning methods and feature optimization in healthcare, ensuring computational efficiency and high predictive reliability. By addressing the challenges of dimensionality reduction and ensuring precise predictions, the proposed model sets a benchmark for machine learning applications in PD diagnosis. The results emphasize the potential of this framework to facilitate early detection and support clinical decision making, with significant implications for advancing AI-driven healthcare systems. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Extra trees classifier Machine learning Parkinson’s disease Recursive feature elimination (RFE)

来源：评论

学校读者我要写书评

暂无评论

A Strengthened Primal-Dual Decomposition Algorithm for Solving Electricity Market Pricing With Revenue-Adequacy and FFR Constraints

IEEE Transactions on Energy Markets, Policy and Regulation

引用

IEEE Transactions on Energy Markets, Policy and Regulation 2024年第3期2卷 379-391页

作者： Goudarzi, Hamed Hesamzadeh, Mohammad Reza Bunn, Derek Fotuhi-Firuzabad, Mahmud Shahidehpour, Mohammad Sharif University of Technology Center of Excellence in Power System Management and Control Tehran11365-11155 Iran Kth Royal Institute of Technology School of Electrical Engineering and Computer Science Stockholm10044 Sweden London Business School Department of Management Science and Operations LondonNW1 4SA United Kingdom Illinois Institute of Technology Electrical and Computer Engineering Department ChicagoIL60616 United States

This paper develops a new decomposition algorithm for solving Electricity Market Pricing (EMP) problem, taking into account both revenue-adequacy and Fast Frequency Reserve (FFR) constraints. Due to revenue-adequacy constraint, a bilevel model of the EMP problem is introduced (BL-EMP). The upper level of the BL-EMP model represents the non-convex unit commitment (UC) decisions as well as the revenue-adequacy constraints of the market participants (generators, loads, and battery-storage owner). The lower level is a convex economic dispatch model with FFR constraint. To tackle the computational complexity of the considered BL-EMP model, this paper develops, tests, and proposes a Strengthened Primal-Dual Decomposition (SPDD) algorithm, which takes benefits from both Benders-like and Lagrange Dual-like algorithms. The new SPDD algorithm has a series of interesting computational properties, which are theoretically discussed in the paper. The SPDD algorithm has better computational performance than standard Benders decomposition algorithm and it also does not need tuning of the Big-M (or disjunctive) parameters for solving the proposed BL-EMP problem. Results from the modified IEEE 24-bus, the IEEE 118-bus, and the IEEE 300-bus system show the superiority of proposed SPDD algorithm over the classic Benders algorithm. © 2023 IEEE.

关键词： Power markets

来源：评论

学校读者我要写书评

暂无评论

An Intelligent Thyroid Diagnosis System Utilizing Multiple Ensemble and Explainable Algorithms with Medical Supported Attributes

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2024年第6期5卷 2840-2855页

作者： Sutradhar, Ananda Al Rafi, Mustahsin Ghosh, Pronab Shamrat, F M Javed Mehedi Moniruzzaman, Md. Ahmed, Kawsar Azad, A.K.M. Bui, Francis M. Chen, Li Moni, Mohammad Ali Daffodil International University Department of Computer Science and Engineering Dhaka1216 Bangladesh Lakehead University Department of Computer Science Thunder BayONP7B 5E1 Canada University of Malaya Department of Computer System and Technology Kuala Lumpur50603 Malaysia Lakehead University Department of Software Engineering Thunder BayONP7B 5E1 Canada University of Saskatchewan Department of Electrical and Computer Engineering SaskatoonSK S7N 5A9 Canada Imam Mohammad Ibn Saud Islamic University Faculty of Science Department of Mathematics and Statistics Riyadh13318 Saudi Arabia Charles Stuart University AI and Cyber Futures Institute BathurstNSW2795 Australia

The widespread impact of thyroid disease and its diagnosis is a challenging task for healthcare experts. The conventional technique for predicting such a vital disease is complex and time-consuming. A data-driven approach may offer predictive solutions, but it relies on all relevant attributes, which are computationally expensive. Hence, we propose a novel machine learning (ML) based disease prediction system that could potentially predict it by considering three crucial steps. First, to reduce the dimension of the dataset, three feature selection techniques were employed, including feature importance (FIS), information gain selections (IGS), and least absolute shrinkage and selection operator (LAS). Moreover, recommended medical references were considered while developing a feature set having the identical attributes as high-risk factors (HRF). Second, the models, including the three stage hybrid classifier (3SHC) and the three stage hybrid artificial neural network (3SHANN), are used as classifiers on the training data set. Third, a local interpretable model-agnostic explanations (LIME) to the 3SHC with the HRF samples was applied to individually explain the predictions. Then, the overall behaviors of both gender and age categories were explored with the help of a partial dependence plot (PDP). Finally, the proposed system is validated with extensive experiments, where the 3SHC achieves an accuracy (ACC) of 99.29%, which can play a crucial role in preventing thyroid disease and alleviating stress in the healthcare sector. © 2020 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Anxiety Prediction: Leveraging Machine Learning with Functional Design Metrics 21

Enhancing Anxiety Prediction: Leveraging Machine Learning wi...

引用

21st International Conference on Electrical engineering, Computing Science and Automatic Control, CCE 2024

作者： Javed, Abqa Ajmal, Sahar Tajammal, Taliah Aslam, Muhammad Martinez-Enriquez, A.M. Sheikh, Usman Sadiq University of Engineering and Technology Department of Computer Science Lahore Pakistan National University of Computer and Emerging Sciences Department of Software Engineering Pakistan Virtual University of Pakistan Department of Computer Science and Information Technology Lahore Pakistan CINVESTAV-IPN D.F. Department of Computer Science Mexico Medicine Bronglais General Hospital Clinical Fellow in Respiratory SY23 1ER United Kingdom

ISBN: (纸本)9798350377545

Fear, nervousness, and social phobia all contribute to anxiety episodes reaching their peak in seconds. Using data collected from patient questionnaires, this study seeks to anticipate and diagnose anxiety disorders effectively and precisely using machine learning approaches. However, current research offers standardized metrics for assessing machine learning approaches throughout the design process. This work presents a systematic method for efficiently extracting literature from digital sources and emphasizes the need for data engineers to evaluate the efficiency of machine learning approaches to assure forecast quality. The International Organization for Standardization (ISO) 9126 standard is used to assess machine learning approaches for anxiety prediction to address this gap. This evaluation considers parameters such as functionality, accuracy, resilience, throughput, explainability, privacy, and correctness. These defined characteristics serve as the benchmark to measure the efficacy of machine learning approaches in predicting anxiety disorders. © 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

Development of a Smart IOT Enabled Arduino Glove Supporting Sign Language Communication 26

Development of a Smart IOT Enabled Arduino Glove Supporting ...

引用

26th International Conference on computer and Information technology, ICCIT 2023

作者： Zaman, Muhammad Shahriar Victor Biswas, Richard Faculty of Science and Technology Department of Computer Science and Engineering Dhaka1229 Bangladesh Faculty of Engineering Department of Electrical and Electronic Engineering 408/1 Kuratoli Dhaka1229 Bangladesh

ISBN: (纸本)9798350359015

Deafness is an unfortunate issue of our society impairing many people. Around 3 million people in Bangladesh suffer from partial or complete deafness. One key challenge for these individuals is communicating with regular people. Various sign languages are available but very few people are knowledgeable about it. Keeping this issue in consideration this project has been developed. The smart Arduino glove allows the user to express their ideas by translating their fingerspelling as well as the gestures into comprehensible messages. The system also allows identification of common sentences by identifying simple gestures. The suggested system if applied in daily life can bring a positive change in the lives of deaf-mute individuals. © 2023 IEEE.

关键词： Accelerometer Alphabets Deaf Digits Fingerspelling Flex sensor Gesture Recognition I2C MPU-6050 Mute PAJ7620 Sign language

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：