检索结果-内蒙古大学图书馆

49th IEEE International Conference on Acoustics, Speech, and Signal processing (ICASSP)

作者： Zhang, Hengxi Shi, Zhendong Hu, Yuanquan Ding, Wenbo Kuruoglu, Ercan E. Zhang, Xiao-Ping Tsinghua Univ Shenzhen Int Grad Sch Shenzhen Peoples R China Ryerson Univ Dept Elect Comp & Biomed Engn Toronto ON Canada

ISBN: (纸本)9798350344868;9798350344851

Quantitative markets are characterized by swift dynamics and abundant uncertainties, making the pursuit of profit-driven stock trading actions inherently challenging. Within this context, Reinforcement learning (RL) - which operates on a reward-centric mechanism for optimal control has surfaced as a potentially effective solution to the intricate financial decision-making conundrums presented. This paper delves into the fusion of two established financial trading strategies, namely the constant proportion portfolio insurance (CPPI) and the time-invariant portfolio protection (TIPP), with the multi-agent deep deterministic policy gradient (MADDPG) framework. As a result, we introduce two novel multi-agent RL (MARL) methods: CPPI-MADDPG and TIPP-MADDPG, tailored for probing strategic trading within quantitative markets. To validate these innovations, we implemented them on a diverse selection of 100 real-market shares. Our empirical findings reveal that the CPPI-MADDPG and TIPP-MADDPG strategies consistently outpace their traditional counterparts, affirming their efficacy in the realm of quantitative trading.

关键词： Quantitative trading multi-agent reinforcement learning constant proportion portfolio insurance time-invariant portfolio protection

来源：评论

学校读者我要写书评

暂无评论

deep learning myocardial infarction segmentation framework from cardiac magnetic resonance images

引用

BIOMEDICAL SIGNAL processing AND CONTROL 2024年 89卷

作者： Al-antari, Mugahed A. Shaaf, Zakarya Farea Jamil, M. Mahadi Abdul Samee, Nagwan Abdel Alkanhel, Reem Talo, Muhammed Al -Huda, Zaid Sejong Univ Coll Software & Convergence Technol Daeyang AI Ctr Dept Artificial Intelligence Seoul 05006 South Korea Univ Tun Hussein Onn Malaysia Fac Elect & Elect Engn Batu Pahat 86400 Johor Malaysia Princess Nourah Bint Abdulrahman Univ Coll Comp & Informat Sci Dept Informat Technol POB 84428 Riyadh 11671 Saudi Arabia Firat Univ Dept Software Engn Elazig Turkiye Southwest Jiaotong Univ Sch Comp & Artificial Intelligence Chengdu Sichuan Peoples R China

Segmentation of myocardial infarction (MI) is a crucial task in the field of heart disease theranostics. Cardiac magnetic resonance imaging (MRI) is a well-known non-invasive imaging technique that provides comprehensive insights into the structure and function of the heart. However, manually interpreting myocardial infarction from multiple MRI frames is time-consuming, labor-intensive, and prone to errors. This study aims to develop an end-to-end deep learning framework that can automatically segment myocardial infarction (MI) and persistent microvascular obstruction (MVO) among the normal tissues of left ventricle (LV), normal myocardium (Myo), and the remaining normal foreground (BG). The proposed framework includes various stages, such as cardiac MR image collection, preprocessing via three enhancement techniques, splitting and training set augmentation, selection of the most suitable artificial intelligence (AI) segmentation model, and performance evaluation and comparison. For the multi-class segmentation process, we adopt and develope four AI state-of-the-art models: UNet, U-Net_VGG16, SegNet, and ResUnet, which are well-regarded for their effectiveness in image segmentation across different computer vision domains. The publicly available benchmark EMIDEC MRI dataset is utilized for training and evaluating the proposed segmentation framework. The ResU-Net achieved the top performance compared to other AI models, recording an overall accuracy (Acc) of 88.48%, recall (Re) of 85.24%, precision (Pre) of 85.46%, F1-score of 85.35, and MIoU of 84.23%. Comparing with the original dataset (without preprocessing), the CLAHE preprocessing improves the ResU-Net segmentation performance by 2.19% and 3.08% in terms of average F1-score and MIoU for all classes (LV, Myo, MI, and MVO). Therefore, the proposed AI segmentation framework demonstrates its potential for effectively performing multi-class segmentation of cardiac diseases from MRI images.

关键词： Cardiac disease Myocardial Infraction (MI) Microvascular obstruction (MVO) AI-based segmentation deep leaning

来源：评论

学校读者我要写书评

暂无评论

deep learning Gesture Recognition System Based on 3D Convolution 9

Deep Learning Gesture Recognition System Based on 3D Convolu...

引用

9th International Conference on Intelligent Computing and Signal processing, ICSP 2024

作者： Long, Ken Li, Zhiyuan Xie, Binghong School of Communications and Information Engineering Chongqing University of Posts and Telecommunications Chongqing400065 China

ISBN: (纸本)9798350376548

The rapid development of the Internet of Things has promoted the progress of human-computer interaction technology, in which gesture recognition, as a key component, provides diversified applications for smart homes, elderly care, and other fields. In recent years, with the wide application of Wi-Fi, gesture recognition based on channel state information (CSI) has attracted much attention due to its advantages of flexibility, convenience, and cost-effectiveness. In this paper, we propose a gesture recognition system that includes an image conversion module and a 3D convolutional network recognition module. In the image conversion module, we treat the CSI subcarrier information from multiple receivers as a one-dimensional time series and convert them into images separately. Then by aggregating all the image data on the new depth dimension, we finally input the processed data into the 3D Convolutional network recognition module for gesture recognition. This module focuses on capturing changes in CSI information over time and space. We evaluated our model on a Widar3.0 open data set, and the experimental results show that the proposed method has a good effect in realizing cross-domain gesture recognition. © 2024 IEEE.

关键词： Wi-Fi

来源：评论

学校读者我要写书评

暂无评论

Analysis of deep Feature Matching Algorithms in UAV Visual Localization

Analysis of Deep Feature Matching Algorithms in UAV Visual L...

引用

2024 International Russian Automation Conference, RusAutoCon 2024

作者： Prutyanov, Viktor V. Ternov, Mikhail A. Kostrov, Daniil S. Hse University Moscow Russia

The work presents an analysis of the application of deep learning-based methods for the keypoint extraction and matching in the context of map-aided UAV visual localization. A method for visual localization in three degrees of freedom is proposed, which employs pretrained SuperPoint and LightGlue networks for both short-term optical flow and global frame-to-map matching components. The ability of SuperPoint and LightGlue to refine a prediction of a less accurate location estimator without fine-tuning was estimated. The accuracy of the proposed method was evaluated and a detailed analysis of keypoint-based image matching algorithms was conducted on the AdM-UAV dataset. The performance evaluation of the proposed method was carried out on NVIDIA Jetson Orin Nano and shows acceptable results for real-time visual localization on this hardware platform. © 2024 IEEE.

关键词： Optical flows

来源：评论

学校读者我要写书评

暂无评论

A Survey on Skin Lesion Detection and Classification using Machine learning 2

A Survey on Skin Lesion Detection and Classification using M...

引用

2nd International Conference on Artificial Intelligence and Machine learning Applications, AIMLA 2024

作者： Yadav, Rashmi Bhat, Aruna Delhi Technological University Department of Computer Science and Engineering Delhi India

ISBN: (纸本)9798350349221

The research explores how dermatologists use machine learning to quickly and accurately identify and classify skin injury. Conventional diagnosis techniques depend on visual examination, but are subjective and have different interpretations. The ability to analyze data and recognize patterns is a possible remedy for ML. The paper examines the current technology of machine learning, focusing on validation in the real world, and deals with data set variability. Although the research acknowledges the fruitfulness of deep learning (DL), the research emphasizes the benefits of traditional ML techniques regarding interpretation and processing performance. Methods for automating the analysis of skin lesions, such as feature engineering, rule-based techniques and traditional ML algorithms, have been studied. The study suggests using advanced transfer learning techniques, integrating genetic and clinical data, and refining the way artificial intelligence (AI) is explained in order to get over barriers. Intending to enhance the accessibility and correctness of skin lesion identification, the future requires collaboration between dermatology and machine learning to develop real-time diagnostic tools. By offering scalable solutions for rapid diagnosis of lesions and improved patient outcomes, this combination of medical expertise and ML capabilities has the power to revolutionize dermatology. The future of automated dermatological diagnosis is expected to be shaped by collaboration between machine learning experts and dermatologists, enabling more personalized treatment for patients. © 2024 IEEE.

关键词： Dermatology

来源：评论

学校读者我要写书评

暂无评论

DePondFi'23 Challenge on real-time Pond Environment: Methods and Results

引用

IEEE ACCESS 2024年 12卷 157975-157987页

作者： Sasithradevi, A. Suganya, R. Prakash, P. Mohamed Mansoor Roomi, S. Vijayalakshmi, M. Nathan, Sabari Kasthuri, P. Persiya, J. Brighty Ebenezer, L. Jain, Sparsh Verma, Sshubam Balasubramanian, S. Sai Subramaniam, M. Sai Sriram, T. Pranav Phanindra Sai, M. Raj, Chandan Yadav, Amandeep Payak, Ritik Paul Choudhury, Suman Singh, Rohit Vellore Inst Technol Ctr Adv Data Sci Chennai 600127 India Vellore Inst Technol Sch Comp Sci & Engn Chennai 600127 India Indian Inst Technol Madras Dept Elect Engn Chennai 600044 India Thiagarajar Coll Engn Dept Elect & Commun Engn Madurai 625015 India Vellore Inst Technol Sch Elect Engn Chennai 600127 India Couger Inc Tokyo Japan Maharaja Agrasen Inst Technol Dept Comp Sci & Engn Delhi 110086 India Sri Sathya Sai Inst Higher Learning Dept Math & Comp Sci Anantapur 515134 Andhra Pradesh India IIT Roorkee Dept Elect Engn Roorkee 247667 India SAHAJ AI Bangalore 560034 India

This article summarizes the Detection of Pond Fish Challenge (DePondfi'23 Challenge), held during the National Conference on Computer Vision, Pattern Recognition, image processing and Graphics (NCVPRIPG 2023). The main goal of the challenge was to find the most effective methods for detecting pond fish in underwater images, overcoming obstacles such as poor visibility, variations in turbidity, and environmental shifts. Sixty participants registered, with 15 teams submitting results for phase 1. The challenge concluded with four teams earning top honors based on mAP (mean Average Precision) score and time complexity. The mAP scores achieved by toppers are as follows: DETECTRON - 38.93%, DMACS SAI - 36.65%, PondVision - 31.63%, and Sahajeevis - 29.06%. This article describes the toppers method and discusses the detection results. Our challenge event is in line with Sustainable Development Goal 14, which focuses on the conservation and sustainable utilization of ponds and marine resources for sustainable development.

关键词： Aquaculture underwater challenges detection models mAP computational time.

来源：评论

学校读者我要写书评

暂无评论

Towards Automated Visual Inspection of Electrical Grid Assets for the Smart Grid - An Application to HV Insulators

Towards Automated Visual Inspection of Electrical Grid Asset...

引用

2024 International Conference on Communications Control and Computing Technologies for Smart Grids

作者： Lopes, Fernando Rocha, Pedro Coelho, Andre Polytech Univ Coimbra RuaMisericerdiaLagar CorticosS. Martinho Bispo P-3045093 Coimbra Portugal Inst TelecomunicaOes Pinhal Marrocos Polo II DEEC P-3030290 Coimbra Portugal EDP Lab Inspecdo Ativos P-2685038 Sacavem Portugal

ISBN: (纸本)9798350318562;9798350318555

The globally growing demand for electric power, projected to double by 2050, requires extensive upgrades in Transport and Distribution (T&D) systems. Despite the need for new infrastructures to address this demand, the reliability of ageing electrical T&D systems remains a more critical concern, where power line insulators account for over 50% of T&D maintenance costs. To boost inspection efficiency, electric utilities are integrating remote inspection technologies into their operations. However, the processing of the growing volume of collected data is currently strongly limited by human interpretation tasks. This research evaluates the performance of computer vision based on deep learning models for automated visual inspection of electrical grid assets, validated with real-world data. The developed work presents an example application of automated visual inspection of HV insulators, detecting defects in disc insulators in visible light images, using two state-of-the-art deep learning models. YOLOv8s achieved a mAP@50 of 87.9%, while Faster R-CNN X101-FPN achieved 87.2% for the same metric. The findings highlight the advantages and limitations of automated visual inspection, enabling utility companies to benefit from higher efficiency inspection processes, reducing the costs and improving the reliability of electrical grid maintenance. Assessing the performance and complexity of data-driven automated visual inspection techniques is crucial for developing streamlined models that effectively handle high data volumes, and that can evolve to real-time operation and integration in the monitoring and control functions of the smart grid or as a dynamic component of a digital twin of the grid.

关键词： Asset Inspection HV Insulator Computer Vision deep learning Smart Grid

来源：评论

学校读者我要写书评

暂无评论

Traffic signs andmarkings recognition based on lightweight convolutional neural network

引用

VISUAL COMPUTER 2024年第2期40卷 559-570页

作者： Zhao, Shuen Gong, Zhikun Zhao, Dongyu Chongqing Jiaotong Univ Sch Mech & Vehicle Engn Chongqing 400074 Peoples R China

Intelligent recognition of traffic signs and markings is an important component of autonomous driving and intelligent transportation systems. It is also an important theoretical basis for autonomous driving path planning. To address the low accuracy and poor real-time performance of the traffic signs and markings detection in complex and multivariate scenes, a lightweight convolutional neural network recognition method in multiple interference scenes is proposed. Firstly, based on Gamma correction and contrast limited histogram equalization algorithm, the traffic signs and markings image under the low illumination condition is enhanced adaptively. Then, the MobileNet-V2 fuse together with deepLab-V3 + algorithm to segment traffic signs and markings in multiple scenes. Finally, an identification model is established to realize the adaptive recognition of traffic signs and markings, which are based on Lightweight Convolutional Neural Networks (Lw-CNN). Besides, the recognition method proposed in this paper is verified by the CSUST Chinese Traffic Sign Detection Benchmark (CCTSDB). The experimental results show that the Mean Intersection Over Union (MIOU) of the MobileNet-V2 and deepLab-V3 + algorithm reaches 83.07%, it increased by 25.8% than before image enhancement. The recognition accuracy of algorithm based on lightweight convolutional neural network is 99.92%, higher than the algorithms of MobileNet-V2 and VGG16.

关键词： Autonomous driving Traffic signs and markings deep learning Lightweight convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis of CNN-based deep learning Models for Landslide Detection

A Comparative Analysis of CNN-based Deep Learning Models for...

引用

2024 Asian Conference on Intelligent Technologies, ACOIT 2024

作者： Oak, Omkar Nazre, Rukmini Naigaonkar, Soham Sawant, Suraj Vaidya, Himadri Department of Computer Science & Engineering COEP Tech University Pune India Department of Computer Science & Engineering Graphic Era Deemed to be University Dehradun India

ISBN: (纸本)9798350374933

Landslides inflict substantial societal and economic damage, underscoring their global significance as recurrent and destructive natural disasters. Recent landslides in northern parts of India and Nepal have caused significant disruption, damaging infrastructure and posing threats to local communities. Convolutional Neural Networks (CNNs), a type of deep learning technique, have shown remarkable success in image processing. Because of their sophisticated architectures, advanced CNN-based models perform better in landslide detection than conventional algorithms. The purpose of this work is to investigate CNNs’ potential in more detail, with an emphasis on comparison of CNN based models for better landslide detection. We compared four traditional semantic segmentation models (U-Net, LinkNet, PSPNet, and FPN) and utilized the ResNet50 backbone encoder to implement them. Moreover, we have experimented with the hyperparameters such as learning rates, batch sizes, and regularization techniques to fine-tune the models. We have computed the confusion matrix for each model and used performance metrics including precision, recall and f1-score to evaluate and compare the deep learning models. According to the experimental results, LinkNet gave the best results among the four models having an Accuracy of 97.49% and a F1-score of 85.7% (with 84.49% precision, 87.07% recall). We have also presented a comprehensive comparison of all pixel-wise confusion matrix results and the time taken to train each model. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Enhancing Helmet Violation Detection and License Plate Recognition through Optimization of YOLOV8 Models with Edge Computing Integration 24

Enhancing Helmet Violation Detection and License Plate Recog...

引用

9th International Conference on Intelligent Information Technology (ICIIT)

作者： Hoa Doan Nguyen Thanh Phu Nguyen Ngoc Thien Nghia Phan Duc Anh Nguyen Phan Tuan Nguyen Le Dinh Tech Univ Ostrava Ostrava Czech Republic FPT Univ Hcm Vietnam

ISBN: (纸本)9798400716713

This paper explores an innovative approach to improve the efficiency of helmet violation detection and license plate recognition through the optimization of YOLOV8 models integrated with edge computing. The proposed method leverages the capabilities of YOLOV8 models to enhance accuracy in identifying helmet violations and recognizing license plates. Additionally, the integration of edge computing facilitates the pre-processing of images at the smart cameras, reducing the data load sent to the central server by approximately 30%. Although there is a slight reduction in accuracy compared to centralized processing, the trade-off is deemed acceptable. The experimental results demonstrate the effectiveness of the optimized system in achieving a balance between accuracy and computational efficiency, thereby presenting a promising solution for real-time monitoring and enforcement of traffic regulations.

关键词： Helmet Detection License Plate Recognition YOLOV8 Edge Computing Smart Camera deep learning

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：