检索结果-内蒙古大学图书馆

Robust Deep learning Model for Black Fungus Detection Based on Gabor Filter and Transfer learning

Computer Systems Science & engineering 2023年第11期47卷 1507-1525页

作者： Esraa Hassan Fatma M.Talaat Samah Adel Samir Abdelrazek Ahsan Aziz Yunyoung Nam Nora El-Rashidy Machine Learning and Information Retrieval Department Faculty of Artificial IntelligenceKafrelsheikh UniversityKafrelsheikhEgypt Electronics and Communication Engineering Department Faculty of Engineering Horus UniversityDamiettaEgypt Information Systems Department Faculty of Computers and InformationMansoura UniversityMansoura35516Egypt Department of ICT Convergence Soonchunhyang UniversityAsanKorea

Black fungus is a rare and dangerous mycology that usually affects the brain and lungs and could be life-threatening in diabetic ***,some COVID-19 survivors,especially those with co-morbid diseases,have been susceptible to black ***,recovered COVID-19 patients should seek medical support when they notice mucormycosis *** paper proposes a novel ensemble deep-learning model that includes three pre-trained models:reset(50),VGG(19),and *** approach is medically intuitive and efficient compared to the traditional deep learning *** image dataset was aggregated from various resources and divided into two classes:a black fungus class and a skin infection *** the best of our knowledge,our study is the first that is concerned with building black fungus detection models based on deep learning *** proposed approach can significantly improve the performance of the classification task and increase the generalization ability of such a binary classification *** to the reported results,it has empirically achieved a sensitivity value of 0.9907,a specificity value of 0.9938,a precision value of 0.9938,and a negative predictive value of 0.9907.

关键词： Black fungus COVID-19 Transfer learning pre-trained models medical image

来源：评论

学校读者我要写书评

暂无评论

The Impact of Blockchain and AI in the Healthcare 3

The Impact of Blockchain and AI in the Healthcare

引用

3rd IEEE International Conference on Electrical Power and Energy Systems, ICEPES 2024

作者： Thatikonda, Ramya Kempanna, M. Tatikonda, Reshmi Bhuvanesh, A. University of the Cumberlands PhD in Information Technology Williamsburg United States Bangalore Institute of Technology Department of Artificial Intelligence and Machine Learning Karnataka Bengaluru India University of the Central Missouri Masters in Computer Information Systems Warrensburg United States PSN College of Engineering and Technology Department of Electrical and Electronics Engineering Tamil Nadu Tirunelveli India

ISBN: (纸本)9798350390728

Integrating blockchain and artificial intelligence (AI) in healthcare presents a transformative opportunity. Therefore, this can revolutionize patient care by improving data management and medical research. Decentralization and invulnerability granted by blockchain technology foster improved data consistency and interoperability. AI enhances blockchain by streamlining data examination, automatic tasks, and decision-making processes, resulting in more precise medical diagnoses and tailored therapy programs. The intersection of blockchain and AI is explored, along with their potential benefits and practical applications in healthcare. The document examines the hurdles encountered when integrating innovations, including scalability, interoperability, and regulatory adherence issues. Additionally, the study investigates moral questions surrounding data ownership and patient consent, thereby protecting confidentiality and trust. The synergy between blockchain, AI, and healthcare will lead to transformative advancements. It pledges to fortify data safety and expedite information sharing. Consequently, this will result in an improved, price-effective, and patient-focused healthcare environment. Recommendations were made for additional research and practical implementation in healthcare technology. The study should concentrate on creating scalable and compatible blockchain systems. A vital factor for extensive use is tackling regulatory and moral issues. Hybrid blockchain designs and federated learning strategies will bolster information privacy and utility. © 2024 IEEE.

关键词： Data consistency

来源：评论

学校读者我要写书评

暂无评论

Avoidance of Ship Collision Using Deep Reinforcement learning with Bi-Directional Long Short-Term Memory in Continuous Action Spaces

Avoidance of Ship Collision Using Deep Reinforcement Learnin...

引用

Data Science and Network Security (ICDSNS), International Conference on

作者： Revatthy Krishnamurthy Myasar Mundher Adnan Sunil Kumar V T. Aditya Sai Srinivas Srinivas Aluvala Artificial Intelligence & Machine Learning Department Rajalakshmi Engineering College Chennai India Department of Computers Techniques Engineering College of Technical Engineering The Islamic university Najaf Iraq Department of Artificial intelligence and Machine learning Nitte Meenakshi Institute of Technology Bengaluru India Artificial Intelligence and Machine Learning Jayaprakash Naryan College of Engineering Mahabubnagar India School of Computer Science and Engineering SR University Warangal India

ISBN: (数字)9798350373110

ISBN: (纸本)9798350373127

Recent days, with the globalization of the world economy the ships count are increasing for marine transportation and the Waterways are growing increasingly overcrowded than previous. This situation may leads to problem of collision of ships which may cause losing of life and damage for property and to nature. Many automatically collision models of ships are implemented but many focused only on the ship-ship encounter situation only. By using a grid sensor which is virtual sensor, agents of Deep Reinforcement learning (DRL) classify approach of a multiple ships. This framework introduces an automatically collision detection algorithm for ships using DRL in continuous action spaces. DRL is used for avoidance of collision with a maximum distance of safe passing between ships. A unique method is developed named inside Obstacle Zone by Target (OZT) used to change learning capability that expands the OZT. Using Bi-directional Long Short-Term Memory (BI-LSTM) cell, network is redesigned and continuous action spaces training is carried out to train a model with longer safe distance of ships. In collision detection model that the bow cross range is effective for COLREGs compliant collision avoidance that is proposed in this model. The propose model also validates a scenario that included more ships and have passed that Imazu problem. The proposed BI-LSTM model achieved 80.43 % of accuracy, 95.67% of precision, 85.89% of recall and 93.54% of f1 score values.

关键词： Accuracy Biological system modeling Bidirectional control Power system stability Deep reinforcement learning Stability analysis Vectors Marine vehicles Collision avoidance Long short term memory

来源：评论

学校读者我要写书评

暂无评论

machine learning-Based Security for Cloud Computing Challenges and Implications 3

Machine Learning-Based Security for Cloud Computing Challeng...

引用

3rd IEEE North Karnataka Subsection Flagship International Conference, NKCon 2024

作者： Dattangire, Rahul Burle, Rushikesh Biradar, Divya Dewangan, Leelkanth Data Engineering Independent Researcher HoustonTX77002 United States Datta Meghe Institute of Higher Education and Research Artificial Intelligence and Machine Learning Faculty of Engineering and Technology Maharashtra Wardha442001 India University of Texas at Arlington Computer Science ArlingtonTX76013 United States G H Raisoni College of Engineering Maharashtra Nagpur India

ISBN: (纸本)9798350364569

Cloud computing"is a computer model that provides end users with quantifiable, scalable, and on-demand services. These days, almost every organization uses computer technology extensively for infrastructure, cost-cutting, advanced platforms, data processing, data analytics, and other uses. End users can use any online application from anywhere to access the cloud service providers' (CSP) services. It is imperative to guarantee the security of cloud infrastructure, and a multitude of research employing various technologies is used to provide more accurate and potent defenses against cloud threats. machine learning has shown to be a more useful technology in recent years for protecting cloud infrastructures. machine learning algorithms are trained to generate models that mechanize the process of recognizing cloud threats more precisely than any other technology by using a range of actual datasets as training grounds. A couple of the most recent research that has worked on machine learning to defend against cloud attacks are examined in this paper. © 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

Blockchain Enabled Secure Data Sharing in Medical Sensors Networks: Concise Review 2

Blockchain Enabled Secure Data Sharing in Medical Sensors Ne...

引用

2nd DMIHER International Conference on artificial intelligence in Healthcare, Education and Industry, IDICAIEI 2024

作者： Raymond, David Kumar, Praveen Gourshettiwar, Palash Werarathna, Induni Mankar, Divyani Verma, Prateek Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Computer Science and Medical Engineering Maharashtra Wardha442001 India Datta Meghe Institute of Higher Education and Research School of Health Allied Sciences Department of Biomedical Sciences Maharashtra Wardha442001 India Datta Meghe Institute of Higher Education and Research Faculty of Engineering and Technology Department of Artificial Intelligence and Machine Learning Maharashtra Wardha442001 India

ISBN: (纸本)9798331528713

Medical sensor networks have transformed patient monitoring and treatment approaches in the ever-changing healthcare industry. However, strong systems for safe and effective data sharing are necessary to fully use sensor data. Conventional methods of exchanging data in healthcare are frequently rife with security flaws, such as the potential for data tampering and privacy violations. Blockchain technology offers decentralized, irreversible, and transparent data sharing and storage capabilities and offers a promising solution to these problems. The potential for safe data sharing offered by Blockchain technology to revolutionize healthcare delivery is examined in this review paper, which examines the convergence of medical sensor networks and Blockchain technology. This paper explores the various Blockchain implementation architectures, approaches, and case studies in healthcare contexts by thoroughly examining current literature. This paper also discusses the prospects, difficulties, and future directions in this emerging subject, emphasizing the effects on interoperability, patient care, and data security. By explaining the revolutionary possibilities of secure data sharing in medical sensor networks afforded by Blockchain technology, the current study hopes to stimulate additional investigation and creativity in the quest for safer and more effective healthcare systems. © 2024 IEEE.

关键词： Block Chain Healthcare Medical Sensor Networks Privacy Secure Data Sharing

来源：评论

学校读者我要写书评

暂无评论

Evaluation of machine learning Models for Kidney Stone Classification Using HOG Features on CT Images

Evaluation of Machine Learning Models for Kidney Stone Class...

引用

2024 International Conference on Intelligent and Innovative Practices in engineering and Management, IIPEM 2024

作者： Ravikumar, S. Chandralekha, E. Vijay, K. Sundar, Koushik Vel Tech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Department of Computer Science & Engineering Chennai India Rajalakshmi Engineering College Department of Artificial Intelligence and Machine Learning Chennai India Citibank NA Jersey CityNJ United States

ISBN: (纸本)9798350390049

Kidney stones, or renal calculi, represent hard mineral and salt deposits which are formed in the kidneys and may cause intense pain, infections of the urinary tracts, and other complications. machine learning techniques are commonly applied to the problem of kidney stone classification, as this task is critical in medical imaging and contributes to correct diagnosis and treatment. This work aims to evaluate the performance of Gradient Boost, XGBoost, AdaBoost, and CatBoost models when processing a set of kidney CT images organized in Cyst, Normal, Stone, and Tumor classes. The initial dataset, available on Kaggle, is placed in the public domain and includes grayscale images resized to 128×128 pixels: these images are further processed using Histograms of Oriented Gradients. It was found that although AdaBoost achieved high precision for Cysts, lower overall accuracy (0.71) and variable performance, particularly in Stone classification, were observed. Conversely, XGBoost, Gradient Boosting, and CatBoost demonstrated near-perfect accuracy (0.99) and ROC values of 1.00 across all categories, with CatBoost showing especially consistent performance. The superior classification capabilities of XGBoost, Gradient Boosting, and CatBoost for kidney stone identification were highlighted, emphasizing the importance of selecting ML models based on detailed performance metrics. Future research may focus on optimizing these models or enhancing feature extraction methods. © 2024 IEEE.

关键词： Medical imaging

来源：评论

学校读者我要写书评

暂无评论

Optimizing the Shopping Experience Through Smart Trolleys with Integrated Deep learning Techniques 4

Optimizing the Shopping Experience Through Smart Trolleys wi...

引用

4th International Conference on Sustainable Expert Systems, ICSES 2024

作者： Akhila, T.S. Presskila, X. Arogya Kalyanakumar, P. Krishnan, R. Santhana Sahanapreethi, M. Sundaravadivel, P. Dhanalakshmi College of Engineering Department of Electronics and Communication Engineering Tambaram Chennai India Scad College of Engineering and Technology Department of Computer Science and Engineering Cheranmahadevi India Velammal College of Engineering and Technology Department of Information Technology Madurai India Scad College of Engineering and Technology Department of Electronics and Communication Engineering Cheranmahadevi India Psna College of Engineering and Technology Department of Information Technology Dindigul India Saveetha Engineering College Department of Artificial Intelligence and Machine Learning Chennai India

ISBN: (纸本)9798331540364

This research study presents an innovative smart shopping trolley system designed to enhance the retail shopping experience by integrating advanced deep learning techniques and sensor technology. The system begins with a secure membership card login process, enabling personalized customer interaction. Upon entry into the trolley system, products can be easily added and identified through a camera utilizing EfficientN et, a deep learning model trained on the store's inventory. The system incorporates an infrared motion sensor and a weight sensor to ensure accuracy in product registration, enhancing reliability and preventing errors. As the shopping process unfolds, a Transformer-based recommendation system analyzes real-time choices and shopping history to provide tailored product suggestions displayed on the trolley's LCD screen. To enhance security, an Autoencoder neural network monitors for anomalies in product identification and removal, alerting store personnel in case of discrepancies. Upon completing the shopping process, a QR code is generated for payment, streamlining the checkout experience. This smart shopping trolley system optimizes the shopping process, leveraging real-time data to improve efficiency, security, and overall customer satisfaction in the retail environment. The system aims to revolutionize traditional shopping methods by applying deep learning and IoT technologies and create a seamless, engaging consumer experience. © 2024 IEEE.

关键词： Customer satisfaction

来源：评论

学校读者我要写书评

暂无评论

Image Copy Move Forgery Detection Using Multi-Plane Convolutional Neural Network 2

Image Copy Move Forgery Detection Using Multi-Plane Convolut...

引用

2nd International Conference on Distributed Computing and Optimization Techniques, ICDCOT 2024

作者： Vagdevi, S. Deepthi, V.S. Dhareshwar, Shweta Al-Farouni, Mohammed Rao, Hanumanthakari Kalyan City Engineering College Artificial Intelligence and Machine Learning Karnataka Bangalore India Karnataka Bangalore India Nitte Meenakshi Institute of Technology Department of Mca Bengaluru India The Islamic University Najaf Iraq Griet Dept. of Cse Telangana Hyderabad India

ISBN: (纸本)9798350382952

In digital images, the digital device usage and availability of open source for image editing applications are easily manipulating the digital images. The Copy Move Forgery (CMF) is an extensive technique for duplicate or hide particular image portion without leaving visual signs. Therefore, it is complex to detect the CMF and the forensic experts are rely on efficient technique for CMF detection. The Multi Plane Convolutional Neural Network (MP-CNN) is proposed in this research for detecting CMF in images. The MICC-F220, MICC-F2000 and CASIA 2.0 datasets are used which is publicly available Kaggle dataset. This dataset is preprocessed by image denoising which reduces the noise from input image. Then, the pre-processed images are provided to MP-CNN for detecting the CMF. The metrics such as precision, accuracy, f1score and recall are applied for estimating the MP-CNN performance. The MP-CNN attains accuracy of 99.45%, 99.26% and 98.61% for MICC-F220, MICC-F2000 and CASIA 2.0 datasets when compared to existing techniques like CNN, Masked Region CNN (M-RCNN) and Contrast Limited Adaptive Histogram Equalization with CNN (CLAHE+CNN). © 2024 IEEE.

关键词： Image denoising

来源：评论

学校读者我要写书评

暂无评论

Modeling and Performance Analysis of PEM Fuel Cells Using machine learning Techniques

Modeling and Performance Analysis of PEM Fuel Cells Using Ma...

引用

Intelligent and Innovative Technologies in Computing, Electrical and Electronics (IITCEE), International Conference

作者： Shamitha Gaddam P Srihith Reddy Shruti V Joshi Department of Artificial Intelligence and Machine Learning BNM Institute of Technology Bengaluru India Department of Electrical Engineering BNM Institute of Technology Bengaluru India

ISBN: (数字)9798331515911

ISBN: (纸本)9798331515928

Proton Exchange Membrane Fuel Cells (PEMFCs) play a vital role in the transition to clean energy technologies, requiring accurate and efficient models for their control, optimization, and real-time monitoring. This research focuses on creating machine learning models as Support Vector Regression (SVR), Gradient Boosting (GBR), XGBoost, and Random Forest (RFR) and MLPRegressor to predict PEMFC key performance attributes. Using a dataset from Hamidi et al [1]., which focuses on Nafion 112 membranes and MEA activation tests. Input features include current density, voltage, power density, gas pressure, relative humidity, and impedance components (real and imaginary). The primary output is cell voltage, a key performance indicator. Among the models, Random Forest Regressor achieved the best results with an accuracy of 94.715, R 2 score of 0.8049, MSE of 0.001036, and RMSE of 0.03219. Feature importance analysis revealed relative humidity, current density, and voltage as the most influential factors. This research illustrates the use of machine learning techniques in modeling the functioning of attributes of PEMFCs, providing insights for optimizing fuel cell performance and accelerating the inculcation of clean energy technologies.

关键词： Accuracy Computational modeling Scalability Fuel cells Humidity Predictive models Real-time systems Current density Random forests Optimization

来源：评论

学校读者我要写书评

暂无评论

Spiralizer Strategy with Wolf Pack Algorithm and Temporal Convolution Network – Gated Recurrent Unit for Coconut Crop Yield Prediction

引用

SN Computer Science 2024年第8期5卷 1108页

作者： Niranjan, S.J. Raviprakash, M.L. Babu, J. Ananda Department of Computer Science and Engineering Kalpataru Institute of Technology Tiptur Affiliated to Visvesvaraya Technological University Karnataka Belagavi India Department of Artificial Intelligence and Machine Learning Kalpataru Institute of Technology Tiptur Affiliated to Visvesvaraya Technological University Karnataka Belagavi India Department of Information Science and Engineering Malnad College of Engineering Hassan Affiliated to Visvesvaraya Technological University Karnataka Belagavi India

The estimating coconut crop yield across large areas is crucial for ensuring food security and early to predicted. However, current spatial heterogeneity learning methods face limitations, such as selecting irrelevant features across regions and difficulties in achieve better performance in crop yields. In this research, Spiralizer Strategy with Wolf Pack Algorithm (SWPA) for feature selection and Temporal Convolution Network – Gated Recurrent Unit (GRU) for classification combined to performed efficiently and achieve better performance. The SWPA efficiently select relevant feature using a spiral search patter, leading to refined feature set that is fed into classification model. The TCN is designed to learn sequential data by applying convolutional layer that maintain the order of sequence. The GRU technique specifically designed to handling sequential data and long term dependencies, reducing the risk of overfitting. Initially,data acquisition and pre-processing involved Min–Max normalization, label encoder which efficiently scaling and categorized values into 0 and 1. The proposed method achiever better outcomes such as better MAE of 1.50, MAPE of 8.05, MSE of 7.23, RMSE of 0.35, R² of 0.654, respectively. The existing method such as Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) are compared to proposed method. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Coconut crop yield predication Convolutional neural network Long short-term memory Spiralizer strategy with wolf pack algorithm and temporal convolution network – gated recurrent unit

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：