检索结果-内蒙古大学图书馆

28th International Conference on Knowledge Based and Intelligent information and Engineering Systems, KES 2024

作者： Alahdal, Nusaybah M. Abukhodair, Felwa Meftah, Leila Haj Cherif, Asma Department of Information Technology King Abdulaziz University Jeddah21589 Saudi Arabia Center of Excellence in Smart Environment Research King Abdulaziz University Jeddah22252 Saudi Arabia PRINCE Research Lab ISITCom H- Sousse Sousse University Sousse4000 Tunisia

AI analytics enables autonomous cars to detect and recognize objects, such as other vehicles, pedestrians, traffic signs, and obstacles, in real-time. deep learning models, notably the You Only Look Once (YOLO) model, have demonstrated accuracy and speed in obstacle avoidance. However, current datasets are limited, lacking diversity and labeling, hindering their ability to represent real-world scenarios accurately. Besides, previous studies have focused extensively on specific object classes, such as pedestrians and vehicles, often neglecting other objects like bikes and road signs. To address this, we introduce a novel dataset tailored for AV environments, encompassing various road object types under different conditions. Our innovative methodology relies on self-supervised learning using the late YOLO version to improve model robustness with limited labeled data and AI-driven adaptive model optimization based on real-time feedback. We evaluate three YOLO architectures-YOLOv5, YOLOv7, and YOLOv8-customized for AV object detection. Our assessment covers everyday AV objects such as cars, pedestrians, bicycles, and road signs, emphasizing early detection. We employ the VSim-AV simulator dataset to ensure robust evaluation, augmented with preprocessing techniques to optimize data quality and model generalization. The study reveals that YOLOv5 and YOLOv8 outperform YOLOv7 regarding precision and recall across various object classes, with YOLOv5 leading at 1.3 ms/image and YOLOv8 at 3.3 ms/image. The mean average precision was 0.94 for YOLOv5, 0.441 for YOLOv7, and 0.927 for YOLOv8, highlighting the limitations in current literature and challenges in YOLO model performance. © 2024 The Authors.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Experimental Evaluation of an Internet of Things enabled Blood Pressure Prediction System using Enhanced learning Methodology 5

Experimental Evaluation of an Internet of Things enabled Blo...

引用

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

作者： Ramkumar, G. Saveetha University Saveetha School of Engineering Department of Ece Chennai India

ISBN: (纸本)9798331504403

There are several needs and immense challenges in the existing health care conditions for which necessitates the development of a new model for blood pressure prediction in an IoT enabled environment. Many traditional models and systems do not have the accuracy or processing speed characteristics that allow real-time analysis of health information. The paper illustrates IoT-based Blood pressure prediction system with the help of Enhanced learning Methodology (ELM) using Hybrid GoogleNet-SVM model. The system reads smart blood pressure monitors and wearable health monitors to stream in live data on key vital stats like systolic and diastolic BP, resting heart rate, and more. Using a middleware platform, the data is transmitted in real-time to a centralized server via CoAP (Constrained Application Protocol) and saved in the cloud platform Azure. Preprocessing steps like data aggregation, timestamp alignment, data cleaning, and data transformation to have an appropriate dataset from all relevant features. The hybrid GoogleNet-SVM model uses both deep learning with GoogleNet to extract complex patterns from the data and for precise blood pressure predictions leveraging the classification power of SVM. Able to deliver highly reliable real-time health advice with an accuracy of 97.56%, the system allows to manage blood pressure proactively and safely. It is a powerful, scalable platform for continuous health monitoring, representing an important leap forward in the field of predictive healthcare technology. © 2024 IEEE.

关键词： Electronic health record

来源：评论

学校读者我要写书评

暂无评论

4K real time image to image Translation Network With Transformers

引用

IEEE ACCESS 2022年 10卷 73057-73067页

作者： Shibasaki, Kei Fukuzaki, Shota Ikehara, Masaaki Keio Univ Fac Sci & Technol Dept Elect & Comp Engn Yokohama Kanagawa 2238522 Japan

CNNs have traditionally been applied in computer vision. Recently, applying Transformer networks, originally a technique in natural language processing, to computer vision has received much attention and produced superior results. However, Transformers and their derivation have drawbacks that the computational cost and memory usage increase rapidly with the image resolution. In this paper, we propose the Laplacian Pyramid Translation Transformer (LPTT) for image to image translation. The Laplacian Pyramid Translation Network, a previous study of this work, creates Laplacian pyramid of the input images and processes each component with CNNs. However, LPTT transforms the high-frequency components with CNNs and the low-frequency components with Axial Transformer blocks. LPTT can have Transformer's expressive power while reducing the computational cost and memory usage. LPTT significantly improves the quality of generated images and inference speed for high-resolution images over conventional methods. LPTT is the first network with a Transformer that can perform practical inference in real time on 4K resolution images. LPTT can also process 8K images in real time depending on the model conditions and the performance of the GPU. The ablation study in this paper suggests that even when processing high-resolution images, the performance is improved while maintaining the inference speed by computing the low-resolution component with a Transformer. LPTT improves PSNR value by 0.41 dB in MIT-Adobe FiveK dataset. The greater the number of layers in the Laplacian pyramid, the greater the improvement of LPTT over the Laplacian Pyramid Translation Network.

关键词： Transformers Laplace equations image resolution Computational efficiency Transforms Task analysis Tensors deep learning image to image translation Laplacian pyramid photo retouching transformer

来源：评论

学校读者我要写书评

暂无评论

Self-supervised pre-training improves fundus image classification for diabetic retinopathy

Self-supervised pre-training improves fundus image classific...

引用

Conference on real-time image processing and deep learning

作者： Lee, Joohyung Lee, Eung-Joo Korea Elect Technol Inst 9fl Elect Ctr 11 Worldcup Bukro 54 Gil Mapo Gu Seoul South Korea MGH Dept Radiol CAMCA Boston MA USA Harvard Med Sch Boston MA 02115 USA

ISBN: (纸本)9781510650817;9781510650800

This paper assesses the efficacy of self-supervised learning in the deepDR Diabetic Retinopathy image Dataset (deepDRiD). Recently, self-supervised learning has achieved great success in the field of Computer Vision. Particularly, self-supervised learning can effectively serve the field of medical imaging where a large amount of labeled data is usually limited. In this paper, we apply the Bootstrap Your Own Latent (BYOL) approach to grade diabetic retinopathy which scores the lowest among the MedMNIST dataset. With the pre-trained model using BYOL, we evaluate the efficacy of the BYOL approach on deepDRiD following fine-tuning protocols. Further, we compare the performance of the model with the model from scratch and proved the effectiveness of BYOL in deepDRiD. Our experiment shows that BYOL can boost the performance of grading diabetic retinopathy.

关键词： Self-supervised learning Diabetic Retinopathy MedMNIST BYOL deep learning

来源：评论

学校读者我要写书评

暂无评论

ARON: Adaptive Resource Optimization Network for AI-Driven Business Management 4

ARON: Adaptive Resource Optimization Network for AI-Driven B...

引用

4th International Conference on Technological Advancements in Computational Sciences, ICTACS 2024

作者： Verma, Pranay Singh, Gurinder Chaudhary, Naina Thakur, Ayush Gupta, Astha Kler, Rajneesh Amity International Business School Amity University Uttar Pradesh Noida India Amity University Tashkent Uzbekistan Amity Institute of Information Technology Amity University Uttar Pradesh Noida India

ISBN: (纸本)9798350387490

This paper introduces the Adaptive Resource Optimization Network (ARON), a novel AI-driven framework for strategic resource allocation and risk management in enterprise environments. ARON integrates deep reinforcement learning, natural language processing, and adaptive learning mechanisms to dynamically adjust resource allocation strategies in real-time, responding to market fluctuations and internal organizational changes. We present the architecture and methodology of ARON, including its data integration module, deep learning core, and natural language interface. An empirical evaluation across three diverse industry sectors-manufacturing, financial services, and e-commerce-demonstrates ARON's superior performance compared to traditional methods and static AI systems. Results show significant improvements in overall performance (up to 28.5%), resource allocation efficiency (13.4% to 40% increase), and risk management (37.2% reduction in Value at Risk). The study highlights ARON's enhanced adaptability to market volatility and its potential to transform business decision-making processes. We discuss the implications of these findings, acknowledge limitations, and propose directions for future research in AIdriven business management. © 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

Research on Soccer Player Tracking Algorithm Based on deep learning 1

引用

3rd EAI International Conference on Application of Big Data, Blockchain, and Internet of Things for Education Informatization, BigIoT-EDU 2023

作者： Bai, Hongding Yuanyuan, Chai Cheng, ZhenHua Yunnan Engineering Vocational College Yunnan Kunming650304 China Sports Department Modern College of Northwest University Shaanxi Xi’an710130 China

ISBN: (数字)9783031631399

ISBN: (纸本)9783031631382

Target tracking technology is of great significance in football game video, and is the basis of high-level semantic tasks such as video summary generation, player motion analysis, game strategy formulation and football event detection. In recent years, many excellent algorithms have emerged in the field of target tracking, mainly including correlation filtering and deep learning, but none of them can achieve high accuracy player tracking for soccer game *** recent years, AI and computer vision have become hot topics, attracted the close attention of a large number of experts and researchers, and triggered an upsurge of extensive and in-depth research. This paper studies the soccer player tracking algorithm based on deep learning. A football player tracking scheme based on deep learning is proposed: a convolutional neural network is built to extract the rich visual features of players in football game video, and the network is trained on a large number of data sets containing similar objects, which improves the ability of the algorithm to identify the same team members. The main objective of the project is to develop a system that can track the position and movement of players in real time, which will be used for live broadcast purposes. In order to achieve this goal, we designed a system using deep learning technology and computer vision algorithm. We also use some advanced technologies, such as 3D graphics processing unit (GPU) and field programmable gate array (FPGA). © ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering 2024.

关键词： Graphics processing unit

来源：评论

学校读者我要写书评

暂无评论

DiffusionBlend: learning 3D image Prior through Position-aware Diffusion Score Blending for 3D Computed Tomography Reconstruction 38

DiffusionBlend: Learning 3D Image Prior through Position-awa...

引用

38th Conference on Neural Information processing Systems, NeurIPS 2024

作者： Song, Bowen Hu, Jason Luo, Zhaoxu Fessler, Jeffrey A. Shen, Liyue Department of Electrical and Computer Engineering University of Michigan Ann ArborMI48109 United States

Diffusion models face significant challenges when employed for large-scale medical image reconstruction in real practice such as 3D Computed Tomography (CT). Due to the demanding memory, time, and data requirements, it is difficult to train a diffusion model directly on the entire volume of high-dimensional data to obtain an efficient 3D diffusion prior. Existing works utilizing diffusion priors on single 2D image-slice with hand-crafted cross-slice regularization would sacrifice the z-axis consistency, which results in severe artifacts along the z-axis. In this work, we propose a novel framework that enables learning the 3D image prior through position-aware 3D-patch diffusion score blending for reconstructing large-scale 3D medical images. To the best of our knowledge, we are the first to utilize a 3D-patch diffusion prior for 3D medical image reconstruction. Extensive experiments on sparse view and limited angle CT reconstruction show that our DiffusionBlend method significantly outperforms previous methods and achieves state-of-the-art performance on real-world CT reconstruction problems with high-dimensional 3D image (i.e., 256 × 256 × 500). Our algorithm also comes with better or comparable computational efficiency than previous state-of-the-art methods. Code is available at: https://***/efzero/DiffusionBlend. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Effective Prediction of Cardiovascular Disease Using deep learning 8th

Effective Prediction of Cardiovascular Disease Using Deep Le...

引用

8th International Conference on Information and Communication Technology for Competitive Strategies, ICTCS 2023

作者： Annabel, L. Sherly Puspha Sruthi, B. Sai Rohini, M. Svetha, B. Sai Department of Artificial Intelligence and Machine Learning St. Joseph’s College of Engineering Chennai600119 India Department of Information Technology St. Joseph’s College of Engineering Chennai600119 India

ISBN: (纸本)9789819702091

Today's leading cause of death worldwide is cardiovascular disease, which has risen to the top of the list of diseases in terms of diagnostic difficulty. Cardiovascular disease is more likely to occur in a person with chest pain, depression, hypertension, smoking, women with early menopause, diabetes, high cholesterol, and over drinking. Early prediction of cardiovascular disease is needed to save more lives. Here comes the saviour Machine learning algorithms that are less expensive with more accuracy. Some of the common machine learning algorithms are implemented to predict the disease. Different techniques provide different accuracies depending on the attributes, dataset, and tools used for implementation. Using the ECG dataset, we create an 11-layer Convolutional Neural Network 2D in this study. We have proposed two models namely Cardiovascular Disease Detection—Machine learning (CVD-ML) that can predict Cardiovascular Disease using real-time numerical data and Cardiovascular Disease Detection—deep learning (CVD-DL) using the ECG image. By using ensembling technique, we have attained the highest accuracy of 94.6% for real-time numerical data and by using Convolutional Neural Network we have attained the accuracy of 99.9% for ECG data. Therefore, Artificial Intelligence techniques used are highly reliable and effective in providing accuracy for cardiovascular disease prediction. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of deep learning Models for image Super-Resolution

A Comparative Study of Deep Learning Models for Image Super-...

引用

2024 Asia Conference on Electronic Technology, ACET 2024

作者： Lim, Jia You Chiew, Yeong Shiong Phan, Raphaël C.-W. Wang, Xin School of Engineering Monash University Malaysia Selangor 47500 Malaysia School of Information Technology Monash University Malaysia Selangor 47500 Malaysia

ISBN: (纸本)9781510681361

Many deep learning-based image super-resolution models exist to effectively up-sample images, with the most notable and reliable architectures being Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), and Generative Adversarial Networks (GANs). To date, model benchmarking has been made only with the same architecture type or only with certain datasets that could potentially be beneficial to the proposed models. In this paper, we present the first-known comparison of state-of-the-art super-resolution models, namely, SwinIR, EDSR, Swin2SR and real-ESRGAN, to serve as a reference baseline for future applications where the modelling complexity, frame rates and overall super-resolution accuracy is of concern. The experiments were conducted by reproducing the models entirely by following the training procedures highlighted in their original paper. Then, we performed the evaluations on the conventional image super-resolution test sets, namely, Set5, Set14, BSD100, Urban100, T91 and Manga109. Our experimental results show that each model has their respective tradeoff between the number of measures taken to suppress the super-resolution artifacts and achieve a higher super-resolution accuracy and the overall model processing times, such as the model convergence speed and their respective frame rates. © 2024 SPIE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

deep Clustering and Transfer learning-Based Anomaly Detection in Thermal Power Plant Control Loops 11th

Deep Clustering and Transfer Learning-Based Anomaly Detectio...

引用

11th Frontier Academic Forum of Electrical Engineering, FAFEE 2024

作者： Xinguang, Liu Baoling, Liu Jun, He Xixi, Liu Yulong, Yuan Xiaocui, Yuan Yongtao, Wang Nanchang Institute of Technology Jiangxi Nanchang330099 China State Grid Jiangxi Electric Power Research Institute Jiangxi Nanchang330096 China

ISBN: (纸本)9789819788194

The implementation of deep learning-based fault diagnosis methodologies has been increasingly observed across diverse sectors within the power industry. This is particularly relevant in contexts where power stations generate vast quantities of operational data that necessitate advanced real-time processing capabilities. However, it has been observed that the efficacy of isolated deep learning models often falls below expectations, primarily due to their limited generalization capabilities. Leading to their limited application in the fault detection of power station automatic control loops. In light of these challenges, the present study introduces an innovative anomaly detection methodology specifically designed for control loops. Based on deep clustering and transfer learning. Initially, the methodology employs an autoencoder clustering algorithm to systematically categorize the operational conditions prevalent in control loops. Subsequently, the VAE-LSTM (Variational Auto Encoder-Long Short Term Memory) model is deployed to meticulously extract the latent features of the difference sequences between controlled parameters and set values. The source domain model undergoes training through the minimization of the loss function, thereby optimizing its parameters for enhanced performance. Finally, by harnessing the principles of transfer learning, the model undergoes fine-tuning of its network parameters by training the feature distribution distance of the LSTM network with target domain data, This process significantly enhances the efficiency and accuracy of fault diagnosis. Experimental validation has confirmed the method's superior performance across datasets from various domains. Moreover, it has demonstrated the capability to conduct real-time fault diagnosis across a multitude of automatic control loops. © Beijing Paike Culture Commu. Co., Ltd. 2025.

关键词： Long short-term memory

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：