检索结果-内蒙古大学图书馆

Effi-CNN: real-time vision-based system for interpretation of sign language using CNN and transfer learning

Multimedia Tools and Applications 2025年第6期84卷 3137-3159页

作者： Pranav Katarya, Rahul Big Data Analytics and Web Intelligence Laboratory Computer Science & Engineering Department Delhi Technological University New Delhi India

The deaf and mute population has difficulty conveying their thoughts and ideas to others. Sign language is their most expressive mode of communication, but the general public is callow of sign language;therefore, the mute and deaf have difficulty communicating with others. A system that can correctly translate sign language motions to speech and vice versa in real time is required to overcome this communication barrier. Effi-CNN, a vision-based Sign Language Recognition (SLR) system employing transfer learning with EfficientNetB2, is proposed in this paper. We have also developed a system covering sign gestures to text in real time. Our approach was evaluated on eight publicly available datasets, including the Massey University gesture dataset, the ArSL2018 dataset, the MNIST-ASL dataset, and others. When comparing our results to state-of-the-art algorithms, the experimental findings showed that our proposed work is more successful than existing models. Our proposed model delivered an accuracy of 99.90 on MNIST dataset. The results show that our Effi-CNN surpasses most currently existing solutions and can categorize many gestures with a low error rate. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

T³Planner: Multi-Phase Planning Across Structure-Constrained Optical, IP, and Routing Topologies

引用

IEEE Journal on Selected Areas in Communications 2025年第5期43卷 1823-1839页

作者： Hao, Yijun Yang, Shusen Li, Fang Zhang, Yifan Zhao, Cong Ren, Xuebin Zhao, Peng Xu, Chenren Wang, Shibo Xi’an Jiaotong University National Engineering Laboratory for Big Data Analytics Xi’an710049 China Xi’an Jiaotong University National Engineering Laboratory for Big Data Analytics Ministry of Education Key Laboratory for Intelligent Networks and Network Security Xi’an710049 China Peking University School of Computer Science Beijing100871 China The Chinese University of Hong Kong Department of Computer Science and Engineering Hong Kong National Engineering Laboratory for Big Data Analytics Xi'an Jiaotong University Xi'an710049 China

Network topology planning is an essential multi-phase process to build and jointly optimize the multi-layer network topologies in wide-area networks (WANs). Most existing practices target single-phase/layer planning, and are incapable of satisfying all rigorous topological structure constraints (e.g., dual-homing rings) defined by network standards and operators, especially in large-scale networks. These significantly limit their usability and performance in production networks. We consider a general topology planning problem with typical structure constraints over three essential phases (greenfield, reconfiguration, and site expansion) and topological layers (optical, IP, and routing topologies). We present, T3Planner, a novel practical solver to this problem in production. Specifically, we develop a structure-driven encoder based on graph neural network (GNN) for concise structure encoding, and design a new learning framework with optical-centric layer compression/reconstruction and rule-aided reinforcement learning (RL) for fast convergence and high performance. Extensive experiments on nine real topologies demonstrate that T3Planner scales to large optical networks with hundreds of sites, saves 46.6% cost, and supports 3.12x more demand when compared to related existing approaches. © 1983-2012 IEEE.

关键词： Network topology

来源：评论

学校读者我要写书评

暂无评论

Energy-Aware Blocking Hybrid Flow Shop Scheduling Problem with Sequence-depend Setup Times and Machine Speed Levels 6

Energy-Aware Blocking Hybrid Flow Shop Scheduling Problem wi...

引用

6th International Conference on Industry 4.0 and Smart Manufacturing, ISM 2024

作者： Missaoui, Ahmed Ozturk, Cemalettin O'Sullivan, Barry Insight Centre for Data Analytics University College Cork Ireland Process Energy and Transport Engineering Munster Technological University Cork Ireland

The industry sector has the second largest energy demand after electricity generation. Given the high trend of energy prices, scarcity of its supply due to political instability and dependency on fossil fuels for its production makes energy the primary challenge for the manufacturing sector to stay competitive. Therefore, in addition to novel equipment technologies that use less energy, energy-efficient scheduling has also been a priority for manufacturing companies. In this study, the Hybrid Flowshop Scheduling Problem with Blocking Constraints and Sequence-Depend Setup Times (BHFS-SDST) is investigated for the minimization of makespan and total energy consumption (TEC). First, a novel bi-objective Mixed Integer Linear Programming (MILP) model is formulated and solved through augmented epsilon constraints. Then, a novel multi-objective approach based on Iterated Greedy meta-heuristic is developed for larger instances. Efficiency and the scalability of the proposed approaches are tested with small, medium, and large instances. Computational experiments show the effectiveness of developed methods in solving the BHFS-SDST problem. © 2025 The Authors.

关键词： Mixed-integer linear programming

来源：评论

学校读者我要写书评

暂无评论

An Attention-Guided Image Super-Resolution for Scene Text 8

An Attention-Guided Image Super-Resolution for Scene Text

引用

8th International Conference on Video and Image Processing, ICVIP 2024

作者： Em, Tithnorakneath Aramvith, Supavadee Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok Thailand Multimedia Data Analytics and Processing Research Unit Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok Thailand

ISBN: (数字)9781510689244

ISBN: (纸本)9781510689237

Scene Text Image Super-Resolution (STISR) has become a significant pre-processing technique to enhance text recognition accuracy in low-resolution scene text images, particularly when addressing the challenges of low-resolution (LR) images in real-world applications. Despite advancements in general and text-focused methods, current techniques often fail to retain the fine-grained and text-specific details necessary for downstream recognition tasks. This paper presents a novel Attention-Guided Scene Text Image Super-Resolution (AG-STISR) framework designed to overcome these challenges. Our approach integrates advanced channel attention and spatial attention mechanisms within a shallow feature extraction module, enabling precise enhancement of key text features such as character edges, contours, and fine structural details. Channel attention helps prioritize the features rich in text-specific details by focusing the model's attention on the channels with the most important data for text recognition. On the other hand, Spatial attention emphasizes locating and highlighting the most significant spatial areas in the image, like character outlines and strokes, ensuring that key elements are maintained while minimizing background distractions. We extensively evaluated our model on the TextZoom dataset and assessed it against various scene text recognition standards. The results demonstrate that our AG-STISR model consistently outperforms state-of-the-art methods, showing significant improvements in PSNR, SSIM, and text recognition accuracy across varying levels of image degradation. This research introduces a compelling solution to STISR, enhancing not only image clarity but also the reliability of text recognition in real-world applications. © 2025 SPIE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

A Distributed Neural Network Architecture for Dynamic Sensor Selection With Application to Bandwidth-Constrained Body-Sensor Networks

引用

IEEE Journal of Biomedical and Health Informatics 2025年第5期29卷 3465-3477页

作者： Strypsteen, Thomas Bertrand, Alexander KU Leuven Department of Electrical Engineering STADIUS Center for Dynamical Systems Signal Processing and Data Analytics LeuvenB-3001 Belgium Leuven.AI KU Leuven Institute for AI LeuvenB-3001 Belgium

We propose a dynamic sensor selection approach for deep neural networks (DNNs), which is able to derive an optimal sensor subset selection for each specific input sample instead of a fixed selection for the entire dataset. This dynamic selection is jointly learned with the task model in an end-to-end way, using the Gumbel-Softmax trick to allow the discrete decisions to be learned through standard backpropagation. We then show how we can use this dynamic selection to increase the lifetime of a wireless sensor network (WSN) by imposing constraints on how often each node is allowed to transmit. We further improve performance by including a dynamic spatial filter that makes the task-DNN more robust against the fact that it now needs to be able to handle a multitude of possible node subsets. Finally, we explain how the selection of the optimal channels can be distributed across the different nodes in a WSN. We validate this method on a use case in the context of body-sensor networks, where we use real electroencephalography (EEG) sensor data to emulate an EEG sensor network for motor execution decoding. For this use case, we demonstrate that the distributed algorithm -with only a small amount of cooperation between the nodes- achieves a performance close to the upper bound defined by a fully centralized dynamic selection (maximum absolute decrease of 4% in accuracy). Furthermore, we observe that our dynamic sensor selection framework can achieve large reductions in transmission energy with a limited cost to the task accuracy, validating it as a practical tool for increasing the lifetime of body-sensor networks. © 2013 IEEE.

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

The Inhibition Activity of Fatty Imidazoline Acetate Derived from Waste Cooking Oil on the AISI 1018 Mild Steel Corrosion

引用

Journal of Bio- and Tribo-Corrosion 2025年第1期11卷 1-11页

作者： Cholidah, Tsana Kusumasari, Lutfiah Lila Febriyanti, Tria Firdaus, Achmad Fadjar Maulana Fuadi, Raihana Putri Nariswari Madurani, Kartika A. Chemistry Department Faculty of Science and Data Analytics Institut Teknologi Sepuluh Nopember Sukolilo East Java Surabaya60111 Indonesia Department of Materials and Metallurgical Engineering Faculty of Industrial Technology and Systems Engineering Institut Teknologi Sepuluh Nopember Sukolilo East Java Surabaya60111 Indonesia

Imidazoline and its derivatives have been proposed as effective organic corrosion inhibitors for AISI 1018 mild steel, though challenges remain regarding their production processes and efficiency in CO2/H2S-saturated brine solutions. In this study, fatty imidazoline acetate (FIA) was synthesized from waste cooking oil and evaluated for its ability to protect AISI 1018 steel in corrosive environments. The chemical composition of the synthesized FIA was confirmed through gas chromatography-mass spectroscopy (GC–MS) and Fourier transform infrared (FTIR) spectroscopy. Corrosion inhibition performance was assessed using potentiodynamic polarization at room temperature, with the most promising results further examined via weight loss measurements and contact angle analyses. A commercial inhibitor commonly used in pipelines with CO2/H2S-saturated brine was also evaluated for comparison. Surface characterization of the steel samples, both with and without inhibitors, was conducted using scanning electron microscopy (SEM). The FIA demonstrated an inhibition efficiency (IE%) of 99.89% at a concentration of 40 mg/L, closely matching the performance of the commercial inhibitor (IE% = 99.96%). This study underscores the potential of FIA as a sustainable and effective corrosion inhibitor in CO2/H2S-saturated environments. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2024.

关键词： Gas chromatography

来源：评论

学校读者我要写书评

暂无评论

Using Blue Whale Technology: An ML Edge Self-Adaptable Vehicle Slowdown Earliest Warning Information System 8th

Using Blue Whale Technology: An ML Edge Self-Adaptable Vehic...

引用

8th International Conference on Microelectronics Electromagnetics and Telecommunications, ICMEET 2023

作者： Murali, N. David, D. Beulah Research Scholar Department of Computer Science and Engineering Saveetha School of Engineering SIMATS Tamilnadu Chennai India Department of Data Analytics Institute of Information Technology Saveetha School of Engineering SIMATS Tamilnadu Chennai India

ISBN: (纸本)9789819784219

Human life is challenged by this main work’s ultimate goal of reducing accidents and ensuring life safety due to the enormous growth of vehicles and those based on safety. Here, cases of suspected drunk driving, reckless driving, etc., are present. In this scenario, vehicles are traveling at full speed and irresponsibly using technology on all roads. An early warning system is used to stop auto accidents. In order to accommodate the city's steadily increasing automobile population, intelligent traffic systems must be designed in a resourceful and sustainable manner using all available technology in both light and dark. The advanced ML's work on unexpected traffic flow and all road conditions resulted in significant real-time changes in the flow of traffic for every significant issue that arose. Here, the Blue Whale technology makes use of clever intelligence to quickly identify, detect, and modify warning systems. The primary objective of the study was to train CNN/BWT on the collected data sets in order to operationalize the unsupervised approach for the prediction of critical situations. Additionally, it makes an effort to clarify the situation so that guests can feel more at ease and help them decide on transfers in time-sensitive situations. IOT is a practical way to locate the nearest object using machine learning methods that are accurate and apply received data sets, resulting in significant findings that support object recognition. The paperwork fully focuses on improving driving knowledge and offering new approaches for quick object recognition. The Fast Warning System Initiative, which aims to increase daytime safety and security, is providing a prior warning system. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Backward wave tubes

来源：评论

学校读者我要写书评

暂无评论

Strategic Blended Learning for Addressing Multiclass Imbalance in Classification

Strategic Blended Learning for Addressing Multiclass Imbalan...

引用

2025 IEEE International Conference on Computational, Communication and Information Technology, ICCCIT 2025

作者： Sridhar, S. Anusuya, S. SRM Institute of Science and Technology Department of CSE Emerging Technologies TN Chennai India Saveetha School of Engineering Department of Data Analytics SIMATS TN Chennai India

ISBN: (纸本)9798331512965

Numerous practical applications involve multiclass classification challenges, where the spread of data across the classes is frequently uneven. Due to this inherent data imbalance, supervised learning models often exhibit a bias toward classifying instances into the category with the highest number of data points, posing a significant problem that requires attention. In the realm of classification of data with multiple classes and imbalanced distributions, researchers in machine learning persistently endeavor to mitigate the classifier's inclination towards dominant class with more sample size. Existing studies aim to alleviate this inclination by alternatively adjusting the data for balance prior to the classifier's learning process or adjusting the classifier's training process to focus more on rare class having least number of samples, or employing hybrid strategies. Existing learning approaches encounter challenges in delineating clear boundaries between samples of classes resulting from an imbalanced class distribution and issues with class overlapping. Consequently, the recognition rate for minority classes is hindered. A novel Strategic Blended Learning (SBL) approach is introduced, employing dual decision trees. The first decision tree is utilized to generate an induced dataset through a grouping approach, while the second decision tree is employed to train and make decisions guided by the induced dataset. A noteworthy aspect of this algorithmic approach is its ability to identify rare data points without affecting the inherent distribution of data. To evaluate the proposed algorithm, various multiclass imbalanced datasets from UCI were employed for classification. The findings revealed that the Strategic Blended Learning method excels in detecting both rare and dominant category samples. © 2025 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Harnessing Medical Big data: Integrating Computational Insights for Enhanced Patient Outcomes 24

Harnessing Medical Big Data: Integrating Computational Insig...

引用

7th International Conference on Digital Medicine and Image Processing, DMIP 2024

作者： Syed Masood, Jafar Ali Ibrahim Vinayagam, Jai Kumar Onteru, Ranjit Kumar Kopparapu, Umamaheswararao Parabathini, Bulah Pushpa Rani Department of IoT School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India Centre for Big Data Analytics QIS College of Engineering and Technology Andhra Pradesh Ongole India

ISBN: (纸本)9798400709586

The rapid proliferation of medical big data has opened unprecedented opportunities for enhancing patient outcomes through advanced computational analysis. This paper explores the integration of big data analytics with clinical practices to deliver personalized medicine, optimize treatment protocols, and improve patient care. Our findings demonstrate that a synergistic approach between computational insights and clinical decision-making can significantly advance healthcare outcomes. This study provides a comprehensive framework for harnessing medical big data, addressing key challenges, and proposing solutions to integrate these insights into everyday clinical practice. Copyright © 2024 held by the owner/author(s). Publication rights licensed to ACM.

关键词： Personalized medicine

来源：评论

学校读者我要写书评

暂无评论

Numerical prediction of ureter stone size using an integrated CFD-ML approach

引用

Neural Computing and Applications 2025年第7期37卷 5325-5341页

作者： Ashraf, Muhammad Mubashar Kamal, Khurram Fahad, Muhammad Noor, N.F.M. Ratlamwala, Tahir Abdul Hussain Islamabad Pakistan College of Science and Engineering Hamad bin Khalifa University Doha Qatar Institute of Mathematical Sciences Faculty of Science Universiti Malaya Kuala Lumpur50603 Malaysia Center for Data Analytics Consultancy & Services Faculty of Science Universiti Malaya Kuala Lumpur50603 Malaysia

Ureteral flow parameters provide significant details about its physical attributes. Ureter is a single transport medium for urine transmission from kidney to ureter and its health is very important for a healthy human body. Understanding the fluid flow behavior can contribute toward the ureter health monitoring including estimation of any kind of blockage in the flow. Using ANSYS Fluent, Computational Fluid Dynamics (CFD) analysis and the grid independence study are carried out through iterative simulation process to achieve the solution independence. The CFD modeling provides tools and techniques to observe varying fluid parameters such as pressure, velocity and effect of the flow on smooth walls. Fluid Structure Interaction (FSI), an effective technique to analyze the effects of such flows on the ureter walls is also employed. Although the exact modeling of the ureter wall is not possible due to its complex physical parameters, some of its available physiological properties can be used to visualize the model of the ureter numerically. The present study is intended to predict the ureter stone size by using the FSI analysis. The simulations are carried out by increasing the stone size gradually from 1.7 to 3.4 mm and the input flow parameters are compared with the output flow parameters within the same solution setup and boundary conditions via artificial neural network in MATLAB. The output results obtained from the FSI simulations are then utilized to generate a prediction model for the ureter stone size. It is observed that the increasing stone size has a significant effect on the ureter wall, causing high stress regions in the point of interaction. The findings also revealed that the predicted size of the ureter stone is the closest to the actual size and with the least mean squared error at 80 optimal neurons. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2024.

关键词： Computational fluid dynamics

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：