检索结果-内蒙古大学图书馆

2024 Asia Pacific Conference on Innovation in Technology, APCIT 2024

作者： Mogal, Amit K. Sonaje, Vaibhav P. Kumar, C.H.N. Santhosh Devi, V. Chithra Gottumukkala, Prasanthi Satheesh Kumar, D. Sandip University School of Computer Science and Engineering Department of Computer Science & Application Maharashtra Nashik India Sandip University School of Computer Science and Engineering Department of Computer Science & Applications Maharashtra Nashik India Anurag Engineering College Department of CSE Telangana Kodada India PSNA College of Engineering and Technology Tamilnadu Dindigul India Gokaraju Rangaraju Institute of Engineering and Technology Department of Information Technology Hyderabad India Hindusthan College of Engineering and Technology Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798350361537

The objective of this study is to examine the effectiveness of a hybrid methodology that combines Long Short-Term Memory (LSTM) and k-Nearest Neighbors (k-NN) models in the context of energy prediction within data centres. The objective of this study is to improve the accuracy of predictions and the efficiency of operations by utilising the combined capabilities of these two machine learning approaches. The proposed LSTM-kNN model has superior performance compared to individual LSTM and k-NN models, as evidenced by a thorough assessment on sample data. It achieves a Mean Absolute Error (MAE) of 9.45, Root Mean Square Error (RMSE) of 12.97, and a coefficient of determination (R2) value of 0.89. The importance of model synergy in enhancing the accuracy of energy forecasts is shown by comparative analysis. The results of this study provide a significant contribution to the advancement of energy management methods in data centres. These findings provide vital insights for stakeholders, enabling them to optimise resource allocation, capacity planning, and sustainability initiatives. Proactive energy optimisation measures and cost-effective operational decisions are examples of practical applications. Potential avenues for future study encompass the exploration of supplementary machine learning techniques and the examination of the influence exerted by external factors on patterns of energy usage. In summary, this work highlights the significance of incorporating machine learning methods to tackle intricate issues in energy prediction and control in data centre settings. © 2024 IEEE.

关键词： k-nearest neighbors

来源：评论

学校读者我要写书评

暂无评论

Traffic Detection and Enhancing Traffic Safety: YOLO V8 Framework and OCR for Violation Detection Using Deep Learning Techniques

Traffic Detection and Enhancing Traffic Safety: YOLO V8 Fram...

引用

2024 International Conference on science, Technology, engineering and Management, ICSTEM 2024

作者： Nagaraj, P. Muthamil Sudar, K. Gurusigaamani, A.M. Sai Vasanth, B. Saranya, V.S.S. Reddy, P. Girish Kumar Department of Computer Science and Engineering Kalasalingam Academy of Research and Education Virudhunagar Krishnankoil India Department of Computer Science and Engineering Mepco Schlenk Engineering College Virudhunagar Sivakasi India

ISBN: (纸本)9798350376913

Making sure traffic is safe and well-managed has become a top priority in the world of contemporary transportation. Using the robust YOLO (You Only Look Once) v8 model in conjunction with Optical Character Recognition (OCR) technology, this research explores the creation and deployment of a state-of-the-art traffic detection system. The main goal is to make roads safer by detecting traffic in real-time and automatically identifying violations. This research is built around the YOLO v8 framework, which is known for its fast and accurate object detection. With YOLO v8 and OCR technology integrated, the system's capabilities are greatly enhanced. Extracting textual information from licence plates allows the system to automatically detect violations like speeding. In order to make it work in a wide variety of international contexts, the OCR component is strong and can handle different fonts, sizes, and languages. To sum up, intelligent traffic management systems have come a long way using the YOLO v8 Framework and optical character recognition (OCR) for violation detection and traffic detection and enhancement. © 2024 IEEE.

关键词： Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Scale-Invariant Feature Transform based Single-Sample Face and Ear Recognition 2

Scale-Invariant Feature Transform based Single-Sample Face a...

引用

2nd IEEE International Conference on IoT, Communication and Automation Technology, ICICAT 2024

作者： Tomar, Vivek Kumar, Nitin Deshmukh, Maroti Singh, Maheep University National Institute of Technology Computer Science & Engineering Department Dehradun India Computer Science & Engineering Department Chandigarh India National Institute of Technology Computer Science & Engineering Department Srinagar Uttarakhand Garhwal India Uttarakhand Doon University National Institute of Technology Computer Science Department Dehradun India

ISBN: (纸本)9798350368109

Recognising a person with just one face or ear image is a challenging sub-problem in the field of biometric recognition. The difficulty arises as parameter estimation becomes challenging from a single image. This article proposes a methodology for recognition from a single face or ear image using scale-invariant feature transform (SIFT) for feature extraction. The proposed method is tested on five public databases, namely ORL, Yale, and AR (for facial images) and AMI and IITD (for ear images). Extensive image processing attacks like scaling (0.1 to 3.0), rotation (30 degrees to 330 degrees), and blurring (0.5 to 4.0) are applied to the test image to evaluate the performance of the method in real-time scenarios. The proposed method achieves excellent classification accuracy and shows robustness towards scaling, rotation, and blurring. The classification of the test image is done using the K-nearest neighbour (KNN) by measuring the Euclidean distance. The classification accuracy results are in the range of 88.86%-96.78% and 73.81%-90.50% for face and ear databases, respectively. The classification accuracy is in the range of 82.38%-96.78% and 81.30%-91.56% against scaling and blurring. Further, the proposed method shows acceptable classification accuracy in the range of 73.81%-95.56% on rotation as well. This proves that feature extraction methods like SIFT can be effective even in difficult scenarios where scarcity or only a few samples are available. © 2024 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Design and Implementation of a Remote Controlled Emulator for Standalone PV Systems 16

Design and Implementation of a Remote Controlled Emulator fo...

引用

16th International Conference on Applied and Theoretical Electricity, ICATE 2024

作者： Calin Comsit, Andreea-Mihaela Cotfas, Daniel Tudor Cotfas, Petru Adrian Transylvania University of Brasov Faculty of Electrical Engineering and Computer Science Brasov Romania

ISBN: (纸本)9798350388107

Immeasurable efforts have been put into strengthening energy security and reducing greenhouse gas emissions by meeting growing energy demand. With declining costs and increasing performance, the deployment of PV systems is growing faster. The focus is on standalone solar photovoltaic systems in rural areas or where grid extension is not feasible. Meteorological conditions impact the standalone PV system's behavior. To avoid critical situations during the PV system operation, it is necessary to implement preventive measures. A PV emulator helps to understand better the behavior of the standalone PV system in critical conditions. This paper presents an intelligent and low-cost PV emulator, with focus on behavior of Li-ion battery, which can be remotely monitored and controlled. The proposed PV emulator consists of: A power supply which has the same behavior as the PV panels, battery storage, consumers which are represented by a electronic load (eload) and remote control functions by using mobile applications. The PV emulator was tested and the behavior of battery was analyzed during the charging and discharging processes. It was determined the capacity of battery is influenced by the values of charging and discharging currents. The ambient temperature influences the internal temperature of Li-ion battery and the PV emulator was stopped when the temperature of Li-ion battery was greater than 53o C. © 2024 IEEE.

关键词： Lithium-ion batteries

来源：评论

学校读者我要写书评

暂无评论

An Efficient Convolutional Neural Network Hardware IP for Epilepsy Detection 17

An Efficient Convolutional Neural Network Hardware IP for Ep...

引用

17th IEEE International Conference on Solid-State and Integrated Circuit Technology, ICSICT 2024

作者： Gong, Yi Zhang, Yuejun Tu, Jiangtao Zou, Rongxin Wen, Liang Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China

ISBN: (纸本)9798350361834

Epilepsy is a common neurological disease which is sudden and unpredictable. In response to the current problems of low efficiency and high false detection rate in epilepsy detection, this paper proposes an efficient convolutional neural network (CNN) hardware intellectual property (IP) design scheme for epilepsy detection. According to the epilepsy pathology, the oversampling technique is used to solve the sample imbalance. Filters out the noise by using band-pass filters, and improves the model by using Z-score normalization. Then, the structure of convolutional neural network is optimized to build an end-to-end network model and train the corresponding weights as well as biases. A single convolutional kernel is used to lighten the network model and reduce the parameters. The 22-channel EEG data are processed in a parallelized manner to fully extract the correlation between the channels. The 3-layer convolutional and max pooling module is designed by combining with pipelining technique. Validated on the CHB-MIT dataset, the epilepsy detection accuracy is reached 97.45%. Under TSMC 65 nm process, the IP core area is measured as 1.625 mm 2 , power consumption is measured as 4.225 mW, single detection delay is 0.01935 ms, and single detection energy consumption is 0.237 μJ/class. © 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

An Efficient Method to Deal with Missing Values and Class Imbalance in Healthcare Data 5th

An Efficient Method to Deal with Missing Values and Class Im...

引用

5th International Conference on Computing, Communications, and Cyber Security, IC4S 2023

作者： Singh, Harpreet Singh, Birmohan Kaur, Manpreet Rani, Suvita Department of Computer Science and Engineering Sant Longowal Institute of Engineering and Technology Sangrur Longowal India Department of Electrical and Instrumentation Engineering Sant Longowal Institute of Engineering and Technology Sangrur Longowal India

ISBN: (纸本)9789819773701

The present era belongs to the age of digital devices, where everything is going to be digitized. This makes the massive production of data at faster rates and brings Big Data to light. The arrival of big data has influenced and reformed many domains like Healthcare, Bioinformatics, Agriculture, etc. Healthcare big data has gained the attention of various researchers as it aids in the early diagnosis and prediction of diseases. Generally, real-world healthcare data contains missingness and imbalance distribution of classes which makes classification a challenging problem. Therefore, an efficient method is introduced to deal with the problems of data incompleteness and class imbalance in healthcare data. A class mean and standard deviation-based imputation strategy is proposed to impute the missing values within the region of a particular class. Further, an Improved Elephant Herding Optimization (IEHO) algorithm is utilized to apply the optimal weights for improving the imbalanced learning in healthcare data. A new fitness function is proposed which considers the data missingness and class imbalance along with the feature weighting. The optimized k-Nearest Neighbor (k-NN) learning algorithm is used to perform the classification. The performance of the proposed method is evaluated on four healthcare datasets: Pima diabetes, Wisconsin Breast Cancer (Original), Hepatitis, and HCC Survival. A comparative analysis with state-of-the-art methods is performed based on four-performance metrics: accuracy, balanced accuracy, g-mean, and f-score. The proposed method obtained higher classification accuracy for Wisconsin Breast Cancer (Original), Hepatitis, and HCC Survival datasets than state-of-the-art methods. Experimental results validate the effectiveness of the proposed method in dealing with the missing values and class imbalance problems in healthcare data. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Multi-granularity feature enhancement network for maritime ship detection

引用

CAAI Transactions on Intelligence Technology 2024年第3期9卷 649-664页

作者： Li Ying Duoqian Miao Zhifei Zhang Hongyun Zhang Witold Pedrycz Department of Computer Science and Technology Tongji UniversityShanghaiChina Project Management Office of China National Scientific Seafloor Observatory Tongji UniversityShanghaiChina Department of Electrical and Computer Engineering AlbertaEdmontonCanada System Research Institute Polish Academy of SciencesWarsawPoland

Due to the characteristics of high resolution and rich texture information,visible light images are widely used for maritime ship ***,these images are suscep-tible to sea fog and ships of different sizes,which can result in missed detections and false alarms,ultimately resulting in lower detection *** address these issues,a novel multi-granularity feature enhancement network,MFENet,which includes a three-way dehazing module(3WDM)and a multi-granularity feature enhancement module(MFEM)is *** 3WDM eliminates sea fog interference by using an image clarity automatic classification algorithm based on three-way decisions and FFA-Net to obtain clear image ***,the MFEM improves the accuracy of detecting ships of different sizes by utilising an improved super-resolution reconstruction con-volutional neural network to enhance the resolution and semantic representation capa-bility of the feature maps from *** results demonstrate that MFENet surpasses the other 15 competing models in terms of the mean Average Pre-cision metric on two benchmark datasets,achieving 96.28%on the McShips dataset and 97.71%on the SeaShips dataset.

关键词： object classification object recognition rough sets rough set theory

来源：评论

学校读者我要写书评

暂无评论

Dependence Induced Representations 60

Dependence Induced Representations

引用

60th Annual Allerton Conference on Communication, Control, and Computing, Allerton 2024

作者： Xu, Xiangxiang Zheng, Lizhong Massachusetts Institute of Technology Department of Electrical Engineering and Computer Science CambridgeMA02139 United States

ISBN: (纸本)9798331541033

We study the problem of learning feature repre-sentations from a pair of random variables, where we focus on the representations that are induced by their dependence. We provide sufficient and necessary conditions for such dependence induced representations, and illustrate their connections to Hirschfeld-Cebelein-Renyi (HGR) maximal correlation functions and minimal sufficient statistics. We characterize a large family of loss functions that can learn dependence induced representations, including cross entropy, hinge loss, and their regularized variants. In particular, we show that the features learned from this family can be expressed as the composition of a loss-dependent function and the maximal correlation function, which reveals a key connection between representations learned from different losses. Our development also gives a statistical interpretation of the neural collapse phenomenon observed in deep classifiers. Finally, we present the learning design based on the feature separation, which allows hyperparameter tuning during inference. © 2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Fault Detection and Classification of Faults in Hydropower plants 5

Fault Detection and Classification of Faults in Hydropower p...

引用

5th International Conference on Smart Electronics and Communication, ICOSEC 2024

作者： Swathi, J.S. Jeffery Austin, T.J. Mounesh, V. Birundha, S. Sri Ramakrishna Engineering College Department of Computer Science and Engineering Coimbatore India

ISBN: (纸本)9798331504403

The hydropower plants are crucial for safeguarding the consistency and the safety of the power generation *** research work presents a extensive study onthe classification of the electrical faults and the fault detection techniques in the hydropower plants. As hydropower systems become more sophisticated, the importance of being able to correctly identify and classify electrical faults has gained ever more significance. The main objective is to devise accurate methods for the identification and classification of the large diversity of electrical faults occurring in hydropower plants. The importance of effective classification and fault detection has increased manifold due to increased complexities of the modern power system coupled with the critical need to ensure reliable operational *** classification approach leverages the advanced machine learning algorithms which includes Support Vector Machine (SVM), Random Forest and Radial Basis Function Network (RBFN) to analyze the electrical signals and patterns associated with the different fault conditions. These techniques would be effective to identify the faults by considering the voltage and current evolved by the different phases of the hydropower plants. Comparisons of accuracy by SVM and Random Forest are made out and thus relate to the classification of fault detection of these hydropower plants. This paper presents a knowledge-based approach for fault diagnosis in hydroelectric power systems. © 2024 IEEE.

关键词： Radial basis function networks

来源：评论

学校读者我要写书评

暂无评论

A System Design for 3D Visualization of Underwater Multi-Target Tracking Based on Unity 3D and GMPHD Filter

A System Design for 3D Visualization of Underwater Multi-Tar...

引用

2025 International Conference on Electronics, Information, and Communication, ICEIC 2025

作者： Kim, Seungho Lee, Scott Uk-Jin Hanyang University Bio-Artificial Intelligence Department of Computer Science and Engineering Ansan Korea Republic of Hanyang University Department of Computer Science and Engineering Ansan Korea Republic of

ISBN: (纸本)9798331510756

This paper proposes a design for a visualization system for underwater multi-target tracking based on the Gaussian Mixture Probability Hypothesis Density (GMPHD) filter using Unity 3D engine. Traditional analysis methods for underwater tracking algorithms rely primarily on two-dimensional graphs and numerical data, making it challenging to intuitively understand the complex behavior of tracking systems in three-dimensional underwater environments. Our proposed system design provides a framework for 3D visualization of target movements, sonar sensor detection areas, and filter estimation results, along with an interactive interface for parameter adjustment. The proposed architecture consists of four main components: simulation core, visualization module, environment simulation, and user interface, which would enable researchers to analyze algorithm performance under various scenarios effectively. Through real-time visualization and parameter adjustment capabilities, the system is designed to offer immediate feedback on how different parameters affect tracking performance. This architectural approach aims to provide researchers with a more intuitive understanding of the GMPHD filter's operation and facilitate the optimization of tracking algorithms for underwater environments. The proposed system design demonstrates the potential of 3D visualization tools in advancing underwater multi-target tracking research and algorithm development. © 2025 IEEE.

关键词： Target tracking

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：