检索结果-内蒙古大学图书馆

5th International Conference on Data Intelligence and Cognitive Informatics, ICDICI 2024

作者： Rajan, S. Palanivel Premkumar, R. Department of Electronics and Communication Engineering Viraganoor Madurai India Kongunadu College of Engineering and Technology Department of Computer Science and Engineering Thottiam Tiruchirappalli India

ISBN: (纸本)9798350389609

This research work presents an attempt to develop a new optimization algorithm known as Hybrid Glowworm Swarm Optimization with Fish Swarm Optimization and Fuzzy Support Vector Machine which can be used for improving the accuracy rate for detection of ovarian cancer based on computer aided diagnosis. The hybrid approach extracts the strong features of both Glowworm Swarm Optimization (GSO) and Fish Swarm Optimization (FSO) on the feature selection process while the Fuzzy Support Vector Machine (FSVM) is used for the classification process. GSO and FSO were used together in order to enhance the search ability and the rate of convergence, to cope with the feature space of ovarian cancer datasets that might be complicated. The FSVM carries this kind of classifications further and addresses inherent uncertainties and overlapping features within data. This integrated method tries to offer more frequent and accurate tool for early detection and may help improve the condition of patients with ovarian cancer. The proposed HGSO-FSO-FSVM framework is yet to be implemented but expected to devise major steps towards enhancement of medical diagnostics especially in early diagnose of ovarian cancer which is crucial in increasing survival rate. © 2024 IEEE.

关键词： Diseases

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Based Underground Sewer Defect Classification Using a Modified RegNet

引用

computers, Materials & Continua 2023年第6期75卷 5451-5469页

作者： Yu Chen Sagar A.S.M.Sharifuzzaman Hangxiang Wang Yanfen Li L.Minh Dang Hyoung-Kyu Song Hyeonjoon Moon Department of Computer Science and Engineering Sejong UniversitySeoul05006Korea Department of Intelligent Mechatronics Engineering Sejong UniversitySeoul05006Korea Department of Information of Communication Engineering and Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea

The sewer system plays an important role in protecting rainfall and treating urban *** to the harsh internal environment and complex structure of the sewer,it is difficult to monitor the sewer *** are developing different methods,such as the Internet of Things and Artificial Intelligence,to monitor and detect the faults in the sewer *** learning is a promising artificial intelligence technology that can effectively identify and classify different sewer system ***,the existing deep learning based solution does not provide high accuracy prediction and the defect class considered for classification is very small,which can affect the robustness of the model in the constraint *** a result,this paper proposes a sewer condition monitoring framework based on deep learning,which can effectively detect and evaluate defects in sewer pipelines with high *** also introduce a large dataset of sewer defects with 20 different defect classes found in the sewer *** study modified the original RegNet model by modifying the squeeze excitation(SE)block and adding the dropout layer and Leaky Rectified Linear Units(LeakyReLU)activation function in the Block structure of RegNet *** study explored different deep learning methods such as RegNet,ResNet50,very deep convolutional networks(VGG),and GoogleNet to train on the sewer defect *** experimental results indicate that the proposed system framework based on the modified-RegNet(RegNet+)model achieves the highest accuracy of 99.5 compared with the commonly used deep learning *** proposed model provides a robust deep learning model that can effectively classify 20 different sewer defects and be utilized in real-world sewer condition monitoring applications.

关键词： Deep learning defect classification underground sewer computer vision convolutional neural network RegNet

来源：评论

学校读者我要写书评

暂无评论

ResNet-50 Powered Masked Face Detection: A Deep Learning Perspective 9

ResNet-50 Powered Masked Face Detection: A Deep Learning Per...

引用

9th IEEE International Conference for Convergence in Technology, I2CT 2024

作者： Choudhary, Nilam Rathore, Pramod Singh Kumar, Loveleen Rajaan, Rajesh Sharma, Aakriti Sinha, Dibakar Swami Keshvanand Institute of Technology Management & Gramothan Department of Computer Science & Engineering Jaipur India Manipal University Jaipur Department of Computer And Communication Engineering Jaipur India Manipal University Jaipur Department of Computer Science And Engineering Jaipur India

ISBN: (纸本)9798350394474

Due to the widespread use of face masks as a result of the COVID-19 pandemic, facial recognition technology, which is routinely employed for security screening in workplaces, is encountering substantial difficulties. These algorithms were commonly trained using photos of unmasked human faces, which posed challenges in recognising individuals wearing masks. In order to tackle this problem, this study introduces a deep learning model that is specifically tailored to properly identify individuals wearing masks. The model utilises the ResNet-50 architecture and has high proficiency in recognising individuals wearing face masks. The findings of this study can be incorporated into existing facial recognition software utilized for security screening purposes. © 2024 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Semi-Supervised Deep Learning Based Method for Abnormality Detection in Videos 1

Semi-Supervised Deep Learning Based Method for Abnormality D...

引用

1st International Conference on Advances in computer science, Electrical, Electronics, and communication Technologies, CE2CT 2025

作者： Chaudhary, Deevesh Kumar, Sunil Dhaka, Vijaypal Singh Manipal University Jaipur Department of Data Science and Engineering Jaipur India Manipal University Jaipur Department of Computer Communication and Engineering Jaipur India

ISBN: (纸本)9798331518578

Anomaly detection refers to recognition of events different from normal ones for example road accident, fight, robbery, arsenal etc. Anomaly identification in real world surveillance videos is an important application of computer vision. The work proposed in the paper detects anomalous events in surveillance videos dataset and is based upon semi supervised deep learning model. We trained the model using UCF Crime dataset that consists of 950 normal videos and 950 anomalous videos. The anomaly videos in the dataset consists of 13 different types of anomalies such as Abuse, arrest, explosion, fight etc. that generally occur in real life. The anomaly in the dataset is labelled at video level and not at a specific frame in a video to define the semi supervised nature of learning paradigm. The extracted 3D features from dataset are fed into the multilayered deep learning model. Experimental results show that our approach has significant improvement over state-of-the-art approaches for anomaly detection in surveillance videos. The accuracy of model comes out to be 83.96 percent, that is improvement over other methods. © 2025 IEEE.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Load Balancing and Energy Efficient Routing in Software-Defined Networking 1st

Load Balancing and Energy Efficient Routing in Software-Defi...

引用

1st International Conference on Advanced Computing, Machine Learning, Robotics and Internet Technologies, AMRIT 2023

作者： Ramkumar, M.P. Rani, J. Lece Elizabeth Jeyarohini, R. Selvan, G. S. R. Emil Konar, Rahul Shiva Department of Computer Science and Engineering Thiagarajar College of Engineering Madurai India Department of Electronics and Communication Engineering PSNA College of Engineering and Technology Dindigul India

ISBN: (纸本)9783031472237

Software-defined networking (SDN), a newly developed architecture that alters the route (paths) of traffic packets, is a flexible and logically centralized control plane that provides a reduction in the network’s energy usage. SDN is designed in such a way that the switches and links can handle a large amount of traffic. Their energy use is not proportional to traffic. Due to the increase in network communication, SDN devices demand much more load and energy. The SDN controller has to be modeled efficiently. The system model of the controller is proposed which efficiently distributes the load. This model helps in traffic routing with minimal energy consumption for networking devices. In addition to the above, it balances the load based on the heterogeneous traffic distribution. The proposed method optimizes the energy consumed by the networking devices by means of using the few active switches, minimizing the number of device state transitions, and handling the volume of traffic. Load balancing is performed for multiple controllers. The switch migration technique is used for migrating the traffic from one controller to the other. Energy-efficient routing is done by the proposed algorithm with reduced device state transition. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2024.

关键词： Software defined networking

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Machine Learning Approach for Electrical Fault Detection and Minimization 2

An Adaptive Machine Learning Approach for Electrical Fault D...

引用

2nd International Conference on Information and communication Technology, ICICT 2024

作者： Biswas, Md. Emran Ahamed, Md Sayem Rafi, Md. Abid Hasan Haque, Md Dulal Department of Electronics and Communication Engineering HSTU Bangladesh Department of Computer Science and Engineering Engineering UIU Bangladesh Shanghai Institute of Technical Physics Chinese Academy of Science China

ISBN: (纸本)9798331508227

In the field of electrical power systems, effective fault detection, and minimization are crucial for maintaining system reliability and operational efficiency. This paper introduces machine learning based Adaptive Fault Reduction Algorithm (AFRA), a novel performance optimization approach designed to reduce fault occurrences significantly of electrical power systems. We used a binary classification dataset with a total of 12,001 data samples for this research. Random Forest (RF) model was used for fault detection and compared its performance against the Artificial Neural Network (ANN) model. RF model demonstrated accuracy, precision, and recall, with scores of 99.79%, 99.64%, and 99.91% respectively, compared to the ANN’s 99.54%, 99.54%, and 99.45%. By integrating AFRA, we reduced the number of fault events from 5,496 to just 6, achieving a 99.89% reduction. Our approach highlights the importance of optimization in fault detection and minimization methodologies, offering a robust framework for improving system reliability. The promising results achieved by AFRA suggest significant potential for its application in real-time monitoring and dynamic fault detection and minimization in power systems. ©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

Integrated normal discriminant analysis in mapreduce for diabetic chronic disease prediction using bivariant deep neural networks

引用

International Journal of Information Technology (Singapore) 2024年第8期16卷 4915-4929页

作者： Ramani, R. Dhinakaran, D. Edwin Raja, S. Thiyagarajan, M. Selvaraj, D. Department of Computer Science and Engineering P.S.R Engineering College Sivakasi India Department of Computer Science and Engineering Vel Tech Rangarajan Dr. Sagunthala R & amp D Institute of Science and Technology Chennai India Department of Electronics and Communication Engineering Panimalar Engineering College Chennai India

This study presents the Normal Discriminant Feature Selection based Regressive Deep Neural MapReduce (NDFS-RDNMR) framework designed for efficient prediction of diabetic chronic diseases using input datasets. The primary aim of NDFS-RDNMR is to enhance accuracy and recall in handling large datasets for chronic disease prediction. The framework integrates the Normal Discriminative Preprocessing Model (NDPM) and bivariant regressive deep artificial neural network with MapReduce (BRDANNMR) classifier. Utilizing the Pima Indian diabetic dataset as input, NDFS-RDNMR conducts feature preprocessing through NDPM to extract relevant features for disease prediction. Non-traditional datasets are transformed into traditional formats via parameter rescaling to fit within predefined value ranges. Min–max normalization is applied to improve system accuracy while preserving data relationships. The BRDANNMR classifier utilizes bivariant regression analysis in the mapping phase to generate intermediary outcomes, which are then classified using a bipolar activation function in the reducer process. The framework achieves high accuracy and recall in early diabetes disease prediction, offering valuable insights for medical practitioners and researchers. © Bharati Vidyapeeth's Institute of computer Applications and Management 2024.

关键词： Artificial neural network Classifier Feature selection MapReduce Normalization Regression analysis

来源：评论

学校读者我要写书评

暂无评论

SPECKLE NOISE DETECTION AND REMOVING BY MACHINE LEARNING ALGORITHMS IN MULTISENSORY IMAGES FOR 5G TRANSMISSION

引用

Scalable Computing 2024年第4期25卷 2679-2686页

作者： DHARANI, M. NAGENDRANATH, M.V.V.S. RAFEE, SHAIK MOHAMMAD KISHORE, G.NAVEEN VENKATAKRISHNAMOORTHY, T. Department of Electronics and Communiation Engineering Mohan Babu University Andhra Pradesh Tirupathi India Department of Computer Science and Engineering Sasi Institute of Technology and Engineering Andhra Pradesh Tadepalligudem India Department of AIML Sasi Institute of Technology and Engineering Andhra Pradesh Tadepalligudem India Department of Electronics and Communication Engineering Andhra Pradesh Tadepalligudem India Department of Electronics and Communication Engineering Sasi Institute of Technology & Engineering AP Tadepalligudem India

The multispectral satellite sensor images have multibands, which have some typical noise. There is difficult to detect this tipical noise with low resolution image. The satellite local or gloval pixel information and quantificationare degraded due to this noise. Many standard transformations and filtering operations are developed for detection and removing of non-gaussion noise, which are not given sophisticated results with existing methods. These statistical characteristics are applied to those samples to identify and quantify present tipical noise. The higher-order statistical based machine learning algorithm is developing to remove the speckle noise from satellite image. In this proposed algorithm, implemented the higher order statistical approache such as skewness, kurtosis based adaptive curvelet methods are implemented for the detection and suppression of speckle noise with retrieve of spectral and spectral values. The proposed algorithm preserves smooth and sharp details and maintains the tradeoff level in multispectral bands is suitable for advanced high speed 5G communication with the effective rate of transmission. The proposed results are verified with suitable statistical parameters such as PSNR,Entropy and ERGAS values. © 2024 SCPE.

关键词： Learning algorithms

来源：评论

学校读者我要写书评

暂无评论

Deep ResNet-Based Classification of Long-Tailed Data with Renyi Entropy as Loss Function 5th

Deep ResNet-Based Classification of Long-Tailed Data with ...

引用

5th International Conference on Computing and Network communications, CoCoNet 2023

作者： Yousuf, Salmaan Kumar, Ranjeet Shah, Shahid Mehraj Electronics & Communication Engineering Department NIT Jammu and Kashmir Srinagar India Computer Science Engineering Department NIT Jammu and Kashmir Srinagar India

ISBN: (纸本)9789819747108

A long-tailed distribution in a dataset is a particular kind of imbalance where data samples from several classes are distributed according to a long-tail distribution. As a result, the predictive ability of deep learning algorithms is heavily biased toward those classes that have high data samples (head classes), which results in subpar performance when classifying minority class (tail classes) data samples. In literature, there are several methods to solve class imbalance problems like resampling inputs and reweighting model parameters. However, these methods suffer heavy overfitting and are hard-to-solve optimization problems during training. The existing technique uses Shannon cross entropy as a loss function. In our approach, we use different loss functions, in particular Renyi entropy, to compare the classification accuracy of ResLT-based deep learning models. The results depict that Renyi entropy loss function outperforms any other loss function when it comes to classification tasks on long-tailed data distributions. We test our proposed approach on the long-tailed version of two standard datasets, Cifar-10 and Cifar-100. We achieved an accuracy of 84.15% on the Cifar-10 and 71.78% on the Cifar-100 dataset, which outperforms any state-of-the-art method for class imbalance tasks. Experimental results demonstrate improvement in the accuracy of residual. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

Towards intelligent and trustworthy task assignments for 5G-enabled industrial communication systems

引用

Digital communications and Networks 2025年第1期11卷 246-255页

作者： Mingfeng Huang Anfeng Liu Neal N.Xiong Athanasios V.Vasilakos School of Computer and Communication Engineering Changsha University of Science&TechnologyChangsha410114China School of Electronic Information Central South UniversityChangsha410083China Department of Computer Science and Mathematics Sul Ross State UniversityAlpineTX 79830USA College of Mathematics and Computer Science Fuzhou UniversityFuzhou350116China Center for AI Research University of AgderGrimstad 999026Norway

With the unprecedented prevalence of Industrial Internet of Things(IIoT)and 5G technology,various applications supported by industrial communication systems have generated exponentially increased processing tasks,which makes task assignment inefficient due to insufficient *** this paper,an Intelligent and Trustworthy task assignment method based on Trust and Social relations(ITTS)is proposed for scenarios with many tasks and few ***,ITTS first makes initial assignments based on trust and social influences,thereby transforming the complex large-scale industrial task assignment of the platform into the small-scale task assignment for each ***,an intelligent Q-decision mechanism based on workers'social relation is proposed,which adopts the first-exploration-then-utilization principle to allocate *** when a worker cannot cope with the assigned tasks,it initiates dynamic worker recruitment,thus effectively solving the worker shortage problem as well as the cold start *** importantly,we consider trust and security issues,and evaluate the trust and social circles of workers by accumulating task feedback,to provide the platform a reference for worker recruitment,thereby creating a high-quality worker ***,extensive simulations demonstrate ITTS outperforms two benchmark methods by increasing task completion rates by 56.49%-61.53%and profit by 42.34%-47.19%.

关键词： Industrial Internet of Things Insufficient workers Trust evaluation Social relation Task assignment

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：