检索结果-内蒙古大学图书馆

2023 Asia-Pacific Conference on Image Processing, Electronics and computers, IPEC 2023

作者： Wang, Peng Gu, Jinguang Institute of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Prov. Key Laboratory of Intelligent Information Processing and Real Time Industrial Systems Wuhan University of Science and Technology Wuhan China

In the case of the specific task of identifying named entities within electronic medical record, it is hard to determine the boundary of nested entities, and existing NER systems have insufficient decoding performance. Based on the pre training model BERT, this paper introduces a novel network structure called Biaffine Layer using a bidirectional LSTM layer. The network uses a dual affine attention mechanism for semantic information learning, which can better interact with the semantic information of entity heads and entity tails, thereby achieving better recognition results for entities. Due to the sparsity of named entity datasets and the uneven distribution of entity categories, traditional binary cross entropy loss functions require multiple rounds of training to decode entities. In this paper, we have modified the binary cross entropy loss function to make the proposed model faster decode the entities that need to be identified. The model performs well, according to the experimental findings. The approach suggested in this paper offers a fresh approach to the NER issue raised by electronic medical records, and it is anticipated to considerably boost the effectiveness and caliber of clinical medical research. © Published under licence by IOP Publishing Ltd.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

DD-Net: Dynamic Network Architecture for Optimized Curve Segmentation and Reduce Computational Redundancy 27th

DD-Net: Dynamic Network Architecture for Optimized Curve Seg...

引用

27th International Conference on Pattern Recognition, ICPR 2024

作者： Cao, Yunxiang Chen, Li Wang, Yubo Feng, Zhida Tian, Jing School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan University of Science and Technology Wuhan China Institute of Systems Science National University of Singapore Singapore Singapore

ISBN: (纸本)9783031783821

Curvilinear Structure segmentation has numerous applications in various fields, including providing a better understanding of defects such as cracks on roads or walls, thereby assessing the structural safety of buildings. Numerous curve segmentation methods based on Convolutional Neural Networks(CNN) have been developed in recent years. However, preserving feature integrity remains a challenge. To address this issue, this paper introduces a double dynamic network, called DD-Net, which consists of a dynamic structure for training and dynamic masking for inference. Firstly, to enhance the nonlinearity of the model, a paradigm to auto-adjust the CNN structure via a trainable module is formed for better fine-grained feature extraction. This dynamic structure (DS) paradigm enables a trade-off to extract or discard the feature data in the model training. Due to the data heterogeneity between the training and inference data, models may contain some useless nodes that are less effective during inference. The dynamic masking (DM) is proposed to omit the useless nodes based on the score difference between the train and inference feature statistics, thereby reducing redundant computations. To further improve the model’s performance on thin-curve segmentation and to preserve feature integrity, we introduce a non-curve suppression (NCS) module. This module focuses on background information while considering foreground prediction to address noisy conditions. The experimental results show that our DD-Net achieves promising results on three benchmark datasets and outperforms state-of-the-art curve segmentation models. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Monkeypox Detection with Deep-Learning and Ensemble of Features: A Study 4

Monkeypox Detection with Deep-Learning and Ensemble of Featu...

引用

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

作者： Rajinikanth, Venkatesan Taniar, David Damaševičius, Robertas Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Division of Research and Innovation Department of Computer Science and Engineering TN Chennai602105 India Monash University Faculty of Information Technology MelbourneVIC3800 Australia Centre of Real Time Computer Systems Kaunas University of Technology Kaunas44249 Lithuania

ISBN: (纸本)9798350387490

There is a growing global prevalence of infectious diseases in the human population and it is imperative to have precise diagnosis and therapy in order to effectively cure these diseases. Monkeypox (Mpox) is an infectious disease that can cause varying degrees of skin infection, ranging from minor to severe. The recent spread of Mpox has been confirmed as Public Health Emergency of International Concern (PHEIC) due to its large infection rate. This study aims to develop a deep-learning (DL) scheme for Mpox identification. To enhance the detection accuracy, this study considered preprocessed skin images for the analysis and employed the binary classification with Ensemble of Deep-Features (EDF) using 5-fold cross validation. The outcome of this investigation validate that the suggested deep learning tool enables the attainment of a detection accuracy of 100% on the augmented images from the Monkeypox Skin Images Dataset (MSID). Compared to the existing results in the literature, this research provides a significant result on the chosen MSID database. © 2024 IEEE.

关键词： Classification Ensemble of deep-features Healthcare Illness Monkeypox

来源：评论

学校读者我要写书评

暂无评论

Brownian-Distribution Algorithm Optimized Efficientnet Features for Normal/Glaucoma Retinal Fundus Image Classification 4

Brownian-Distribution Algorithm Optimized Efficientnet Featu...

引用

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

作者： Rajinikanth, Venkatesan Damaševičius, Robertas Taniar, David Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Division of Research and Innovation Department of Computer Science and Engineering TN Chennai602105 India Kaunas University of Technology Centre of Real Time Computer Systems Kaunas44249 Lithuania Monash University Faculty of Information Technolog MelbourneVIC3800 Australia

ISBN: (纸本)9798350387490

The eye is a critical sensory organ in human physiology, and any abnormality in the eye will result in vision problems ranging from mild to severe. Image-guided methods are frequently employed to conduct clinical-level eye health assessments, and retinal fundus imaging (RFI) is among the most frequently employed modalities for eye analysis. This research aims to implement a Deep-Learning (DL)-based technique for classifying the Normal/Glaucoma RFI database. This scheme consist following phases, image collection and labelling, feature extraction with EfficientNet, feature optimization using a novel Brownian-Distribution Algorithm (BDA), and binary classification with 3-fold cross validation. Initially, the proposed approach performs the classification with EfficientNet-model features and documents the results. Subsequently, the BDA is employed to optimize the features in order to identify the most significant ones. Serial features-fusion is then implemented to generate a new features vector. The results of this study confirm that this approach is capable of achieving a 100 % accuracy rate when the BDA optimized feature vector is utilized for the classification task. In the future, this application may be utilized to analyze the RFI collected from hospital. © 2024 IEEE.

关键词： Ophthalmology

来源：评论

学校读者我要写书评

暂无评论

Femoropopliteal Artery Stent Detection in Fluoroscopy Using Otsu's Preprocessing and CNN Segmentation 4

Femoropopliteal Artery Stent Detection in Fluoroscopy Using ...

引用

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

ISBN: (纸本)9798350387490

computer algorithm supported medical image analysis is a prevalent method in clinics, which effectively alleviates the diagnostic burden associated with traditional image assessment methods. This study aims to create a method that utilizes a Convolutional-Neural-Network (CNN) to accurately segment the Femoropopliteal Artery Stent (FAS) in Digital Fluoroscopy Images (DFI). In order to improve the accuracy of segmentation, this study used the combination of Otsu's tri-level thresholding and the Lévy-Flight Distribution Algorithm (LFDA). The suggested technique consists of many stages: collecting and resizing images and masks, boosting the DFI using Otsu and LFDA, using CNN-segmentation to extract the stent region, comparing segmented stent with mask, and calculating the required performance metrics. The efficacy of the suggested system is validated by analyzing the raw and thresholded DFI. The investigational results of this work affirm that this method achieves a higher level of accuracy (>96%) in segmenting images when the Otsu's preprocessed image is utilized. The experimental study employs the UNet-variants, and in the future, alternative CNN-segmentation methodologies can be utilized to validate the effectiveness of the presented scheme. © 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 Open Source Simulation Framework for Moving-Network-Convoy Based Cellular V2X Communication in Intelligent Traffic systems

An Open Source Simulation Framework for Moving-Network-Convo...

引用

IEEE Conference on Vehicular Technology (VTC)

作者： Venkatnarayanan Lakshminarasimhan Alois Knoll Chair of Robotics Artificial Intelligence and Real-Time Systems Department of Computer Engineering School of Computation Information and Technology Technical University of Munich Munich Germany

Open source system level simulators play a crucial role in the early stage research and optimization of cellular V2X communication networks and cooperative intelligent transportation systems by providing for accurate virtual modelling, rapid evaluation of system level design and algorithm development, and by offering the possibility to build-up on previous research. Amidst the various architectures under consideration for future cellular V2X, small moving cells or moving network convoys are promising candidates, considering their inherent mobility. The orchestration of such architectures in combination with the availability of precise information about the vehicles from road side infrastructure open up new possibilities for a cost-optimal deployment and a resource-optimal operation of the cellular V2X networks. Towards this goal, we propose and present an open source simulation framework for moving-network-convoy based cellular V2X applications, and make it available for wider use at https://***/tum-air/ConvoyV2X

关键词：

来源：评论

学校读者我要写书评

暂无评论

First Mile: An Open Innovation Lab for Infrastructure-Assisted Cooperative Intelligent Transportation systems

First Mile: An Open Innovation Lab for Infrastructure-Assist...

引用

IEEE Symposium on Intelligent Vehicle

作者： Rui Song Andreas Festag Abhishek Dinkar Jagtap Maximilian Bialdyga Zhiran Yan Maximilian Otte Sanath Tiptur Sadashivaiah Alois Knoll Fraunhofer Institute for Transportation and Infrastructure Systems IVI Ingolstadt Germany Chair of Robotics Artificial Intelligence and Real-Time Systems Technical University of Munich Garching Germany Technische Hochschule Ingolstadt Ingolstadt Germany

ISBN: (数字)9798350348811

ISBN: (纸本)9798350348828

Infrastructure-assisted Cooperative Intelligent Transportation systems (C-ITS) leverage roadside intelligent infrastructure and vehicular network technology to facilitate information exchange among traffic participants, enhancing road safety, efficiency, and sustainability. However, this requires not only the massive deployment of infrastructure, including advanced sensors, communication devices, and computing units at various levels, but also the involvement of various stakeholders in the development of functions and real-road testing. In this paper, we present our test field – First Mile, as an open innovation lab for C-ITS in Ingolstadt, Germany, offering an open environment for world-wide stakeholders to conduct research and testing in C-ITS. In particular, we equip a 3.5 km area with 22 roadside intelligent masts and 89 sensors, achieving dense deployment on public roads. Our design includes a protocol stack tailored for different C-ITS stations and services, conforming to European communication standards. Furthermore, we conduct quantitative analyzes of key performance metrics, such as radio signal quality and End-to-End delay, to assess the efficacy of First Mile in different data processing pipelines. Finally, we delve into the future prospects of large-scale C-ITS deployment, guided by extensive and prolonged measurement studies.

关键词： Measurement Technological innovation Sensor phenomena and characterization Road safety Stakeholders Sustainable development Intelligent sensors

来源：评论

学校读者我要写书评

暂无评论

Automatic Platform Configuration and Software Integration for Software-Defined Vehicles

arXiv

引用

arXiv 2024年

作者： Pan, Fengjunjie Lin, Jianjie Rickert, Markus Robotics Artificial Intelligence and Real-Time Systems School of Computation Information and Technology Technical University of Munich Munich Germany Multimodal Intelligent Interaction Faculty of Information Systems and Applied Computer Sciences University of Bamberg Bamberg Germany

In the automotive industry, platform configuration and software integration are mostly manual tasks performed during the development phase, requiring consideration of various safety and non-safety requirements. This manual process often leads to prolonged development cycles and provides limited flexibility. This paper introduces a novel approach to automate platform configuration and software integration for software-defined vehicles (SDVs), shifting these activities from the development phase to runtime. Our approach features an integration manager that combines model-based methods and virtualization technologies to generate and execute deployment plans. By leveraging model-based systems engineering (MBSE), our method automatically generates platform configuration and software integration plans, which are then converted into deployment-ready formats using code generation techniques. Utilizing virtualization and container orchestration technologies, the proposed system enables dynamic and flexible resource allocation while ensuring compliance with safety requirements. Communication between the development and runtime platforms is facilitated via a REST API. A proof of concept was implemented on a simulated SDV platform with the Intel Whiskey Lake Board. This demonstration showcases the integration manager on an SDV with a central computer, highlighting the potential to shorten development cycles and adapt to diverse vehicle configurations. © 2024, CC BY.

关键词： Automotive industry

来源：评论

学校读者我要写书评

暂无评论

Cost-Effective Network Packet Manipulation and Error-Detection Device

Cost-Effective Network Packet Manipulation and Error-Detecti...

引用

Consumer Communications and Networking Conference, CCNC IEEE

作者： Alexander F. K. Janus Steffen Moser Frank Slomka Faculty of Engineering Computer Science and Psychology Institute of Embedded System/Real-Time Systems School of Advanced Professional Studies Ulm University Ulm Germany

real-time multimedia communication applications like video conferencing and screen sharing tools are an integral part of our daily communications. These time-critical services are handled via the Internet, a packet-switched and essentially best-effort network, not explicitly designed for robust real-time communication. From a video conferencing service provider's perspective, network issues cause a lot of user complaints due to various reasons: lossy WLAN channels, poorly configured home routers, or crowded carrier-grade network address translation devices. All of these issues have in common that they are often time-variant and non-deterministic. Furthermore, the end-to-end transmission in these highly inhomogeneous networks often suffers from a mixture of more than one error source. To enable a detailed and impartial analysis of multimedia service quality, it is necessary to analyze packets in defective online scenarios and introduce a broadly similar error pattern in an otherwise stable lab network. In this paper, we present SEDER, a hardware-software design of a layer two transparent Ethernet bridge that can be used for two purposes: error detection and error recreation. Implemented on low-cost hardware, the bridge can be easily deployed anywhere to allow debugging, even at a user's site. It is possible to parameterize our bridge to imitate almost all possible network effects, including complex combinations of more than one effect. In addition, we present two case studies, firstly emphasizing the capabilities of SEDER and secondly showing the need to use SEDER to harden open source Video Conferencing Tools against errors.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generating Manipulation Sequences using Reinforcement Learning and Behavior Trees for Peg-In-Hole Task

Generating Manipulation Sequences using Reinforcement Learni...

引用

2022 IEEE International Conference on systems, Man, and Cybernetics, SMC 2022

作者： Xu, Jiahua Lin, Yunhan Zhou, Haotian Min, Huasong Wuhan University of Science and Technology College of Computer Science and Technology Wuhan430065 China Hubei Province Key Laboratory of Intelligent Information Processing and Real-time Industrial System Wuhan430065 China Institute of Robotics and Intelligent Systems Wuhan University of Science and Technology Wuhan430081 China

ISBN: (数字)9781665452588

ISBN: (纸本)9781665452588

Reinforcement Learning (RL), a method of learning skills through trial-and-error, has been successfully used in many robotics applications in recent years. We combine manipulation primitives (MPs), behavior trees (BTs), and reinforcement learning to propose an algorithm for peg-in-hole tasks, which speeds up the convergence of the RL model and enhance the adaptability of the dynamic environment. Manipulation primitives are used as actions for RL, which can reduce the gap between control instruction and robotic actions and speed up the convergence of the RL model. Behavior trees are used as robot behavior control, which makes robots can actively adapt to the changes in the environment. In experiments, RL-BT, from the combination of RL and BT, is designed for the peg-in-hole task in the Gazebo simulation environment by using UR5 as the actuator. The experiments are conducted on a simple peg and a complex multi-hole peg by three aspects, which include convergence speed verification experiment, adaptability of dynamic environment experiment, and algorithm robustness experiment. The experiment result proves that our RL-BT can speed up the convergence and adapt to the changes in the environment. © 2022 IEEE.

关键词： Reinforcement 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

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

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：