检索结果-内蒙古大学图书馆

A deep learning fusion model for accurate classification of brain tumours in Magnetic Resonance images

CAAI Transactions on Intelligence technology 2024年第4期9卷 790-804页

作者： Nechirvan Asaad Zebari Chira Nadheef Mohammed Dilovan Asaad Zebari Mazin Abed Mohammed Diyar Qader Zeebaree Haydar Abdulameer Marhoon Karrar Hameed Abdulkareem Seifedine Kadry Wattana Viriyasitavat Jan Nedoma Radek Martinek Department of Information Technology Lebanese French UniversityErbilIraq Department of Computer Science University of ZakhoZakhoKurdistan RegionIraq Department of Computer Science College of ScienceNawroz UniversityDuhokKurdistan RegionIraq Department of Artificial Intelligence College of Computer Science and Information TechnologyUniversity of AnbarRamadiIraq Department of Telecommunications VSB‐Technical University of OstravaOstravaCzech Republic Department of Cybernetics and Biomedical Engineering VSB‐Technical University of OstravaOstravaCzech Republic Department of Computer Network and Information Security Technical College of Informatics-AkreDuhok Polytechnic UniversityDuhokIraq Information and Communication Technology Research Group Scientific Research CenterAl‐Ayen UniversityThi‐QarIraq College of Computer Sciences and Information Technology University of KerbalaKarbalaIraq College of Agriculture Al‐Muthanna UniversitySamawahIraq Department of Applied Data Science Noroff University CollegeKristiansandNorway Faculty of Commerce and Accountancy Chulalongkorn Business SchoolChulalongkorn UniversityBangkokThailand

Detecting brain tumours is complex due to the natural variation in their location, shape, and intensity in images. While having accurate detection and segmentation of brain tumours would be beneficial, current methods still need to solve this problem despite the numerous available approaches. Precise analysis of Magnetic Resonance Imaging (MRI) is crucial for detecting, segmenting, and classifying brain tumours in medical diagnostics. Magnetic Resonance Imaging is a vital component in medical diagnosis, and it requires precise, efficient, careful, efficient, and reliable image analysis techniques. The authors developed a Deep Learning (DL) fusion model to classify brain tumours reliably. Deep Learning models require large amounts of training data to achieve good results, so the researchers utilised data augmentation techniques to increase the dataset size for training models. VGG16, ResNet50, and convolutional deep belief networks networks extracted deep features from MRI images. Softmax was used as the classifier, and the training set was supplemented with intentionally created MRI images of brain tumours in addition to the genuine ones. The features of two DL models were combined in the proposed model to generate a fusion model, which significantly increased classification accuracy. An openly accessible dataset from the internet was used to test the model's performance, and the experimental results showed that the proposed fusion model achieved a classification accuracy of 98.98%. Finally, the results were compared with existing methods, and the proposed model outperformed them significantly.

关键词： brain tumour deep learning feature fusion model MRI images multi‐classification

来源：评论

学校读者我要写书评

暂无评论

Enhancing Mobile Multimedia Trustworthiness through Federated AI-based Content Authentication: Enhancing Mobile Multimedia

引用

Journal of Mobile Multimedia 2023年第6期19卷 1415-1438页

作者： Rajesh, M. Vengatesan, K. Sitharthan, R. Dhanabalan, Shanmuga Sundar Gawali, Mahendra Bhatu Department of Computer Engineering Sanjivani College of Engineering India Centre for Smart Grid Technologies School of Electrical Engineering Vellore Institute of Technology Chennai India Functional Materials and Microsystems Research Group RMIT University Melbourne 3001 VIC Australia Department of Information Technology Sanjivani College of Engineering India

The rapid proliferation of mobile devices and multimedia content has led to an increased need for ensuring trustworthiness and authentication of the shared data. Traditional centralized methods have proven to be insufficient in maintaining privacy and addressing scalability issues. This paper presents a novel approach to enhancing mobile multimedia trustworthiness through the application of Federated AI-based content authentication techniques. By leveraging the benefits of distributed machine learning and edge computing, our proposed framework efficiently authenticates multimedia data while preserving user privacy and reducing latency. Our system employs a federated learning model that trains AI algorithms on local devices, allowing them to collaboratively build a robust and accurate authentication model. Additionally, this research introduces a blockchain-based decentralized trust management system to further enhance the integrity and traceability of the authentication process. Through extensive evaluations, this research demonstrate that our proposed framework significantly improves the trustworthiness of mobile multimedia content while minimizing the overhead and resource consumption associated with traditional centralized approaches. © 2023 River Publishers.

关键词： content authentication decentralized trust management Federated AI mobile multimedia trustworthiness

来源：评论

学校读者我要写书评

暂无评论

MARS: Multi-Agent Deep Reinforcement Learning for Real-Time Workflow Scheduling in Hybrid Clouds with Privacy Protection 30

MARS: Multi-Agent Deep Reinforcement Learning for Real-Time ...

引用

30th IEEE International Conference on Parallel and Distributed Systems, ICPADS 2024

作者： Cheng, Long He, Haoyang Gu, Yan Liu, Qingzhi Zhao, Zhiming Fang, Fang North China Electric Power University State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources School of Control and Computer Engineering Beijing China Wageningen University and Research Information Technology Group Netherlands University of Amsterdam Research Group of Multi-Scale Networked Systems Netherlands

ISBN: (纸本)9798331515966

Scheduling workflows in hybrid cloud environments presents significant challenges due to the inherent complexity of workflows and the dynamic nature of cloud resources. This complexity is further increased when attempting to balance workflow performance with privacy protection. Recent efforts have leveraged deep reinforcement learning (DRL) to address these challenges. However, most of these approaches rely on single-agent models, which can lead to security issues and scalability problems due to their centralized processing. Specifically, the properties of workflows are transferred to the single agent, which risks leaking privacy information. Our paper addresses these issues by introducing MARS, a real-time workflow scheduling method that prioritizes privacy protection in hybrid clouds. MARS leverages multi-agent deep reinforcement learning (MADRL) to optimize the workflow scheduling of cloud virtual machines (VMs). The benefit of our solution is that it relies on the collaborative learning of multi-agents on multiple VMs, which could assign user data to specific cloud servers for privacy protection while sharing training experiences between agents. In our implementation, MARS aims to reduce workflow completion time and operational costs while complying with strict privacy protection guidelines. The experimental results demonstrate that MARS can significantly surpass existing methods, reducing makespan by an average of 53.18% and costs by 61.98% compared to basic techniques, and achieving 20.26% and 25.71% improvements over the latest advanced methods, respectively. © 2024 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Plant Disease Detection and Classification Using Transfer Learning

Plant Disease Detection and Classification Using Transfer Le...

引用

2024 International Conference on Innovative Computing, Intelligent Communication and Smart Electrical Systems, ICSES 2024

作者： Karthigaiveni, S. Madhubala, C. Avinash, G. Saranya, E. Rohini, P. Yogapriya, G. Chennai Institute of Technology Department of Electronics and Communication Engineering Chennai India Vel Tech Multi Tech Dr. Rangarajan Dr. Sakunthala Engineering College Department of Information Technology Avadi Tamilnadu Chennai India School of Environment and Architecture Design Srm Institute of Science and Technology Tamilnadu Chennai India Sri Eshwar College of Engineering Department of Computer Science Engineering Chennai India Saranathan College of Engineering Department of Computer Science and Engineering Tamilnadu Trichy India Chettinad School of Architecture Chettinad Academy of Research and Education Tamilnadu Chennai India

ISBN: (纸本)9798331543617

Agriculture is a fundamental component of human civilization. It contributes to the economy while also providing sustenance. Plant foliage or crops are susceptible to many illnesses during agricultural agriculture. The illnesses impede the development of their respective species. Timely and accurate identification and categorization of illnesses may mitigate the risk of further harm to the plants. The identification and categorization of these disorders have emerged as significant challenges. The conventional methods used by farmers to forecast and identify plant leaf diseases may be tedious and inaccurate. Challenges may occur while endeavoring to manually forecast illness kinds. The failure to promptly identify and categorize plant diseases may lead to the devastation of crops, causing a substantial reduction in yield. Farmers using automated image processing techniques in their fields may mitigate losses and enhance productivity. This study proposes using a deep learning framework utilizing the latest convolutional neural network model EfficientNet to identify tomato illnesses based on 18,161 plain and segmented photos of tomato leaves. The efficacy of two segmentation theories, namely U-net and Modified U-net, for leaf segmentation is shown. The proposed methodology for plant disease identification is straightforward and computationally efficient, necessitating less time for predictions compared to previous deep learning methodologies. This document presents the computed accuracies for several plant and leaf diseases. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Three-dimensional reconstruction of mobile binocular stereo vision based on push-broom line structured light for a workpiece surface

引用

Journal of the Optical Society of America A: Optics and Image Science, and Vision 2023年第11期40卷 2096-2107页

作者： Wang, Yue Han, Xueyou Rui, Jing Zhang, Hailan Yin, Lei Zhang, Xuefeng Wang, Xiangjun School of Computer Science and Technology Anhui University of Technology Maanshan China Anhui Engineering Research Center for Intelligent Applications and Security of Industrial Internet Anhui University of Technology Maanshan243032 China School of Electrical and Information Engineering Anhui University of Technology Maanshan China The 53th Research Institute of China Electronics Technology Group Corporation Tianjin China State Key Laboratory of Precision Measuring Technology and Instruments Tianjin University Tianjin China Key Laboratory ofMOEMS of the Ministry of Education Tianjin University Tianjin China

Stereo vision technology based on line structured light can effectively solve the problem of a three-dimensional (3D) reconstruction of a smooth surface. A method for 3D reconstruction of mobile binocular stereo vision based on push-broom line structured light for a workpiece surface is presented in this paper. The subpixel coordinates of the light strip centers of the line structured light are obtained by the Steger algorithm while the binocular module moves along the guide rail, and the polar constraint is used to achieve the matching of the extracted light strip centers. As a result, the3Dcoordinates of the light strip centers in each location can be calculated because of the known interior and external parameters of the binocular module. To obtain the 3D point cloud data of the entire surface, a relative pose optimization method with respect to the initial frame is proposed, which accurately estimates the pose of the cameras in each location with respect to that in the initial location and unifies the 3D coordinates of the light strip centers in each location to the datum coordinates. The relative pose optimization method first estimates the rough values by using the direct linear transform method, and then iteratively calculates the refined solutions based on the principle of minimizing the re-projection errors. Simulation data and substantial experimental results validate the effectiveness of our method. Our method is compared to the direct linear transform method and the frame-by-frame transfer method, and the root mean square error (RMSE) of the distance from 3D point cloud to fitted plane is used to evaluate the 3D reconstruction accuracy. The repeatability experiment shows that the RMSE fromour method is as lowas 0.83mm. © 2023 Optica Publishing group.

关键词： Location

来源：评论

学校读者我要写书评

暂无评论

Fuzzy Difference Equations in Diagnoses of Glaucoma from Retinal Images Using Deep Learning

引用

computer Modeling in engineering & Sciences 2024年第4期139卷 801-816页

作者： D.Dorathy Prema Kavitha L.Francis Raj Sandeep Kautish Abdulaziz S.Almazyad Karam M.Sallam Ali Wagdy Mohamed Department of Mathematics Voorhees CollegeVelloreTamil NaduIndia Chandigarh Group of Colleges JhanjeriMohaliPunjabIndia Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityP.O.Box 51178Riyadh11543Saudi Arabia Faculty of Science and Technology School of IT and SystemsUniversity of CanberraCanberraAustralia Operations Research Department Faculty of Graduate Studies for Statistical ResearchCairo UniversityGiza12613Egypt Applied Science Research Center Applied Science Private UniversityAmman11937Jordan

The intuitive fuzzy set has found important application in decision-making and machine *** enrich and utilize the intuitive fuzzy set,this study designed and developed a deep neural network-based glaucoma eye detection using fuzzy difference equations in the domain where the retinal images *** image detections are categorized as normal eye recognition,suspected glaucomatous eye recognition,and glaucomatous eye *** degrees associated with weighted values are calculated to determine the level of concentration between the fuzzy partition and the retinal *** proposed model was used to diagnose glaucoma using retinal images and involved utilizing the Convolutional Neural Network(CNN)and deep learning to identify the fuzzy weighted regularization between *** methodology was used to clarify the input images and make them adequate for the process of glaucoma *** objective of this study was to propose a novel approach to the early diagnosis of glaucoma using the Fuzzy Expert System(FES)and Fuzzy differential equation(FDE).The intensities of the different regions in the images and their respective peak levels were *** the peak regions were identified,the recurrence relationships among those peaks were then *** partitioning was done due to varying degrees of similar and dissimilar concentrations in the *** and dissimilar concentration levels and spatial frequency generated a threshold image from the combined fuzzy matrix and *** distinguished between a normal and abnormal eye condition,thus detecting patients with glaucomatous eyes.

关键词： Convolutional Neural Network(CNN) glaucomatous eyes fuzzy difference equation intuitive fuzzy sets image segmentation retinal images

来源：评论

学校读者我要写书评

暂无评论

Novel heuristics for Stock portfolio optimization using machine learning and Modern Portfolio Theory 2

Novel heuristics for Stock portfolio optimization using mach...

引用

2nd International Conference on Business Analytics for technology and Security, ICBATS 2023

作者： Joemon, Bristo Ghazanfar, Mustansar Ali Azam, Muhammad Awais Jhanjhi, N.Z. Khan, Arshad Ali University of East London Department of Engineering & Computing London United Kingdom School of Information Technology Technology Innovation Research Group Whitecliffe New Zealand SCS Taylor's University Malaysia School of Computer Science Malaysia Elm Research & Innovation ELM Riyadh Saudi Arabia

ISBN: (纸本)9798350335644

Stock market portfolio optimization is a very important aspect for stock market trading. Various techniques have been proposed in the literature ranging from technical indicator-based methods to statistical methods and machine learning for portfolio optimization. However, there has been lack of effort to find the impact of machine learning along with technical indicators to build optimal stock portfolios for trading. From this line of research, this paper focusses on the use of deep learning-based LSTM (Long Short Term Memory) models along with Ichimoku cloud indicators to build optimal portfolios, which can be used in a practical stock market environment. This model is developed with combining the above prediction models such as LSTM and Ichimoku Cloud Indicators with Modern Portfolio Theory and Sharpe Ratio to optimize the model by reducing the risk to reward ratio. The model is tested with real life data of London Stock Exchange and Bombay Stock Exchange. By doing extensive experimental analysis, it was found that, the model combining LSTM network and Ichimoku Indicator, paired with Modern Portfolio Theory and Sharpe Ratio giving consistent performance in the market despite volatility. © 2023 IEEE.

关键词： Long short-term memory

来源：评论

学校读者我要写书评

暂无评论

Bit-Interleaved Polar Coded Modulation with Iterative Successive Cancellation List Decoding

引用

China Communications 2022年第5期19卷 54-68页

作者： Hang Mu Zheng Ma George K.Karagiannidis Panagiotis D.Diamantoulakis The 10th Research Institute of China Electronics Technology Group Corporation Chengdu 610036SichuanChina the School of Information Science and Technology Southwest Jiaotong UniversityChengdu 610036SichuanChina the Departmentof Electrical and Computer Engineering Aristotle University of ThessalonikiThessaloniki 54124Greece

This paper introduces bit-interleaved polar coded modulation with iterative detection/decoding(BIPCM-ID).In order to enable the soft output of successive cancellation list(SCL)decoding,two types of re-encoders are proposed,namely the max-re-encoder and min-re-encoder,*** the iterative decoding,we analytically verify that the average mutual information(AMI)between the transmitted and decoded symbols can approach the Shannon bound with the proposed ***,bit error rate(BER)and block error rate(BLER)in single-user and multi-user scenarios are ***,simulation results show that the performance of BIPCMID outperforms other bit-interleaved coded modulation(BICM)systems with LDPC and Turbo codes,while also reducing the computational complexity.

关键词： BIPCM-ID Polar code SCMA Reencoder SCL AMI

来源：评论

学校读者我要写书评

暂无评论

Malicious traffic detection method based on field feature deep learning

Malicious traffic detection method based on field feature de...

引用

2024 International Conference on computer Application and information Security, ICCAIS 2024

作者： Chen, Zhouguo Deng, Chen Li, Xinze Gao, Xiang Mao, Jiayue Hu, Hangyu School of Computer Science and Engineering Southeast University Jiangsu Nanjing211100 China The 30th Research Institute China Electronics Technology Group Corporation Chengdu610041 China School of Information and Communication Engineering University of Electronic Science and Technology of China Chengdu611731 China The Affiliated Shenzhen School of Guangdong Experimental High School Shenzhen518100 China

ISBN: (纸本)9781510689312

With the rapid development of cloud computing services and security technologies, more and more enterprises are embracing deep binding of business with cloud services, but at the same time, malicious attacks targeting cloud services are gradually becoming rampant. Since the traffic in the cloud service log contains more comprehensive and rich semantic information, malicious traffic detection can be seen as the anomalous text classification task. Therefore, this paper proposes a novel detection method based on deep learning of field features. Firstly, the word vectorization preprocessing is carried out through field selection, decoding, and character encoding, and then the improved TextCNN (Convolutional Neural Networks for Sentence Classification) is used to extract the field features from multiple dimensions. Finally, the classification is carried out. Experimental results show that, compared with the classical models SVM and RF (Random Forest), our proposed method achieves better classification results in malicious cloud service traffic detection tasks. © The Authors.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Game Theory-Based IoT Efficient Power Control in Cognitive UAV

引用

computers, Materials & Continua 2022年第7期72卷 1561-1578页

作者： Fadhil Mukhlif Norafida Ithnin Omar B.Abdulghafoor Faiz Alotaibi Nourah Saad Alotaibi Information Assurance and Security Research Group(IASRG) School of ComputingFaculty of EngineeringUniversiti Teknologi MalaysiaJohorMalaysia Electronic and Telecommunication Department College of EngineeringThe American University of KurdistanIraq Faculty of Computer Science and Information Technology Universiti Putra MalaysiaSelangorMalaysia Computer Department Faculty of Applied CollegeImam Abdulrahman Bin Faisal UniversityDammamSaudi Arabia

With the help of network densification,network coverage as well as the throughput can be improved via ultra-dense networks(UDNs).In tandem,Unmanned Aerial Vehicle(UAV)communications have recently garnered much attention because of their high agility as well as widespread *** this paper,a cognitive UAV is proposed for wireless nodes power pertaining to the IoT ground ***,the UAV is included in the IoT system as the source of power for the wireless nodes as well as for resource *** quality of service(QoS)related to the cognitive node was considered as a utility function based on pricing scheme that was modelled as a non-cooperative game theory in order to maximise users’net utility ***,an energy efficiency non-cooperative game theory power allocation with pricing scheme(EE-NGPAP)is proposed to obtain an efficient power control within IoT wireless ***,uniqueness and existence of the Nash equilibrium have been demonstrated mathematically and through *** results show that the proposed energy harvest algorithm demonstrated considerable decrease in transmitted power consumption in terms of average power reduction,which is regarded to be apt with the 5Gnetworks’***,the proposed algorithm requires around 4 iterations only to converge to NE which makes the algorithm more suitable in practical heterogeneous scenarios.

关键词： UAV drones WSN IoT game theory energy efficiency 5G&B5G networks

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：