检索结果-内蒙古大学图书馆

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Liu, Hui Jia, Chen Cheng, Xu Liu, Xiufeng Shi, Fan School of Computer Science and Engineering Tianjin University of Technology Tianjin China Department of Technology Management and Economics Technical University of Denmark Kongens Lyngby Denmark

ISBN: (纸本)9798350368741

Achieving pixel-level crack segmentation in complex scenarios is a major challenge, as current methods have difficulty effectively integrating both local features and irregular pixel dependencies. In this paper, we introduce a Cascaded Fusion Network (LICFN) specifically designed for crack segmentation, which extracts fine local details and pixel dependencies using a hybrid feature extractor and effectively enhances and fuses them through a cascaded fusion module. To comprehensively evaluate the network, we also created a benchmark dataset, TUT, which includes various scenarios. Experimental results show that our method surpasses others, achieving F1 and mIoU scores of 0.8439 and 0.8509, respectively. The dataset is available at https://***/Karl1109/TUT. © 2025 IEEE.

关键词： cascaded fusion crack dataset irregular dependencies local patterns structural cracks segmentation

来源：评论

学校读者我要写书评

暂无评论

Adaptive Vehicle Detection in Urban Environments: A Self-learning Approach 15th

Adaptive Vehicle Detection in Urban Environments: A Self-le...

引用

15th International Symposium on Ambient Intelligence, ISAmI 2024

作者： Guerrero-Contreras, Gabriel Balderas-Díaz, Sara García-Pascual, Abel Muñoz, Andrés Department of Computer Science and Engineering University of Cadiz Av. Universidad de Cádiz 10 Puerto Real Cádiz11519 Spain

ISBN: (纸本)9783031831164

With increasing urbanization, efficient urban traffic management is a critical challenge that requires smarter and more adaptable systems. This paper introduces a self-learning algorithm designed to enhance the adaptability and effectiveness of vehicle detection models using urban camera infrastructures. By leveraging these ubiquitous devices, the study aims to capture and analyze real-time traffic data, a task traditionally limited by the need for extensive manual data labeling and the limitations of pre-trained models under varying urban conditions. Our self-learning algorithm addresses these challenges by reducing reliance on manual labeling and enabling continuous model adaptation to new conditions without direct human intervention. Implemented in the dynamic urban environment of the city of Madrid, Spain, this study evaluates the algorithm’s capacity to enhance vehicle detection, considering a diverse range of vehicle types. The core of the algorithm comprises an iterative self-training process that refines model performance using both labeled and unlabeled data, thus progressively enhancing detection accuracy. Our findings reveal significant improvements in the ability of the model to accurately identify and classify vehicles, highlighting the potential of self-learning algorithms in urban traffic management. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

Swept-Source Optical Coherence Tomography at Green Wavelengths Using Broadband Frequency Doubling of NIR MEMS-VCSELs 29

Swept-Source Optical Coherence Tomography at Green Wavelengt...

引用

Optical Coherence Tomography and Coherence Domain Optical Methods in Biomedicine XXIX 2025

作者： Jayaraman, Vijaysekhar Burgner, Chris Higgins, Grace Siedschlag, S.J. Hwang, Yunchan Fujimoto, James G. Praevium Research Inc Santa BarbaraCA93117 United States Department of Electrical Engineering and Computer Science Research Laboratory of Electronics Massachusetts Institute of Technology CambridgeMA02139 United States

ISBN: (纸本)9781510683587

Visible wavelength optical coherence tomography (vis-OCT) enables high-resolution retinal oximetry in the 500-600nm wavelength range, through spectroscopic measurements of total hemoglobin bound to oxygen (HbO2) relative to total hemoglobin. Current research in vis-OCT uses spectral domain OCT (SD-OCT) with broadband supercontinuum sources. Although these systems afford high axial resolution, they have significant limitations in terms of sensitivity, imaging range, axial scan rate, and commercial scalability. Swept source OCT (SS-OCT) systems using near infrared (NIR) micro-electromechanical systems tunable vertical cavity surface emitting lasers (MEMS-VCSELs) have achieved MHz A-scan rates and meter scale imaging ranges. No MEMS-VCSELs or broadband swept wavelength sources, however, have been demonstrated in 500-600nm range. In this paper, we demonstrate a visible swept source using broadband frequency doubling of an amplified 1050nm MEMS-VCSEL in a thin-film periodically poled lithium niobate (PPLN) waveguide. We show a usable tuning range of 518-544nm, SS-OCT axial resolution in air/tissue of 6.95/5.25 um, >10mm image range, and >5mW peak power over the tuning range. Our PPLN waveguide uses a linearly chirped poling period to broaden the doubling bandwidth, creating strong spectral modulation on the frequency doubled light. Since the spectral modulation is repeatable, and since green wavelength fringes remain phase-continuous across spectral dips, the spectral modulation can be numerically compensated. Some artifacts remain, however, highlighting the need to flatten the spectrum. Additionally, we demonstrate the first MEMS-VCSELs covering 1049-1184nm, corresponding to a more optimal doubled wavelength range of 524.5-592nm. © 2025 SPIE.

关键词： Optical coherence tomography

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Framework with Learning without Forgetting for Intelligent Surveillance in IoT-enabled Home Environments in Smart Cities

Recent Advances in Computer Science and Communications

引用

Recent Advances in computer science and Communications 2025年

作者： Dalal, Surjeet Dahiya, Neeraj Verma, Amit Faujdar, Neetu Rathee, Sarita Jaglan, Vivek Rani, Uma Le, Dac-Nhuong Department of Computer Science & Engineering Amity University Haryana Gurugram India Department of Computer Science & Engineering SRM University Sonipat Haryana Delhi-NCR India Department of Computer Science and Engineering Maharaja Agrasen University Baddi India Department of Computer Science & Applications GLA University UP Mathura India Department of Electronics and Communication Engineering School of Engineering and Technology JECRC University Rajasthan Jaipur303905 India Department of Computer Science & Engineering Amity University M.P. Gwalior India Department of Computer Science & Engineering K. R. Mangalam University Haryana Gurugram India Dean of Faculty of Information Technology Haiphong University Viet Nam

Background: Internet of Things (IoT) technology in smart urban homes has revolutionised sophisticated monitoring. This progress uses interconnected devices and systems to improve security, resource management, and resident safety. Smart cities use technology to improve efficiency, sustainability, and quality. Internet of Things-enabled intelligent monitoring technologies are key to this goal. Objectives: Intelligent monitoring in IoT-enabled homes in smart cities improves security, convenience, and quality of life from advanced technologies. Using live monitoring and risk identification tools to quickly discover and resolve security breaches and suspicious activity to protect citizens. Intelligent devices allow homeowners to remotely control lighting, security locks, and surveillance cameras. Using advanced technologies to regulate heating, cooling, and lighting based on occupancy and usage. Method: This study introduces a deep learning architecture that uses LwF (Learning without Forgetting) to keep patterns while absorbing new data. The authors use IoT devices to collect and analyse data in real-time for monitoring and surveillance. They use sophisticated data pre-processing to handle IoT devices' massive data. The authors train the deep learning model with historical and real-time data and cross-validation to ensure resilience. Result: The proposed model has been validated on two different Robloflow datasets of 7382 images. The proposed model gains an accuracy level of 98.27%. The proposed Yolo-LwF model outperforms both the original Yolo and LwF models in terms of detection speed and adaptive learning. Conclusion: By raising the bar for intelligent monitoring solutions in smart cities, the suggested system is ideal for real-time, adaptive surveillance in IoT-enabled households. By embracing adaptability and knowledge retention, authors envision heightened security and safety levels in urban settings. © 2024 Bentham science Publishers.

关键词： Smart homes

来源：评论

学校读者我要写书评

暂无评论

UltraSegNet:A Hybrid Deep Learning Framework for Enhanced Breast Cancer Segmentation and Classification on Ultrasound Images

引用

computers, Materials & Continua 2025年第5期83卷 3303-3333页

作者： Suhaila Abuowaida Hamza Abu Owida Deema Mohammed Alsekait Nawaf Alshdaifat Diaa Salama Abd Elminaam Mohammad Alshinwan Department of Computer Science Faculty of Prince Al-Hussein Bin Abdallah II for Information TechnologyAl Al-Bayt UniversityMafraq25113Jordan Medical Engineering Department Faculty of EngineeringAl-Ahliyya Amman UniversityAmman19328Jordan Department of Information Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh11671Saudi Arabia Faculty of Information Technology Applied Science Private UniversityAmman11931Jordan Faculty of Computers Science Misr International UniversityCairo11800Egypt Jadara Research Center Jadara UniversityIrbid21110Jordan

Segmenting a breast ultrasound image is still challenging due to the presence of speckle noise,dependency on the operator,and the variation of image *** paper presents the UltraSegNet architecture that addresses these challenges through three key technical innovations:This work adds three things:(1)a changed ResNet-50 backbone with sequential 3×3 convolutions to keep fine anatomical details that are needed for finding lesion boundaries;(2)a computationally efficient regional attention mechanism that works on high-resolution features without using a transformer’s extra memory;and(3)an adaptive feature fusion strategy that changes local and global featuresbasedonhowthe image isbeing *** evaluation on two distinct datasets demonstrates UltraSegNet’s superior performance:On the BUSI dataset,it obtains a precision of 0.915,a recall of 0.908,and an F1 score of *** the UDAIT dataset,it achieves robust performance across the board,with a precision of 0.901 and recall of ***,these improvements are achieved at clinically feasible computation times,taking 235 ms per image on standard GPU ***,UltraSegNet does amazingly well on difficult small lesions(less than 10 mm),achieving a detection accuracy of *** is a huge improvement over traditional methods that have a hard time with small-scale features,as standard models can only achieve 0.63–0.71 *** improvement in small lesion detection is particularly crucial for early-stage breast cancer *** from this work demonstrate that UltraSegNet can be practically deployable in clinical workflows to improve breast cancer screening accuracy.

关键词： Breast cancer ultrasound image segmentation classification deep learning

来源：评论

学校读者我要写书评

暂无评论

Hybrid Machine Learning Models for Intrusion Detection in IoT: Leveraging a Real-World IoT Dataset

arXiv

引用

arXiv 2025年

作者： Akif, Md Ahnaf Butun, Ismail Williams, Andre Mahgoub, Imadeldin Department of Electrical Engineering and Computer Science Florida Atlantic University Boca RatonFL United States

Cincinnati, · OH The rapid growth of the Internet of Things (IoT) has revolutionized industries, enabling unprecedented connectivity and functionality. However, this expansion also increases vulnerabilities, exposing IoT networks to increasingly sophisticated cyberattacks. Intrusion Detection Systems (IDS) are crucial for mitigating these threats, and recent advancements in Machine Learning (ML) offer promising avenues for improvement. This research explores a hybrid approach, combining several standalone ML models such as Random Forest (RF), XGBoost, K-Nearest Neighbors (KNN), and AdaBoost, in a voting-based hybrid classifier for effective IoT intrusion detection. This ensemble method leverages the strengths of individual algorithms to enhance accuracy and address challenges related to data complexity and scalability. Using the widely-cited IoT-23 dataset, a prominent benchmark in IoT cybersecurity research, we evaluate our hybrid classifiers for both binary and multi-class intrusion detection problems, ensuring a fair comparison with existing literature. Results demonstrate that our proposed hybrid models, designed for robustness and scalability, outperform standalone approaches in IoT environments. This work contributes to the development of advanced, intelligent IDS frameworks capable of addressing evolving cyber threats. © 2025, CC BY-NC-SA.

关键词： Cybersecurity

来源：评论

学校读者我要写书评

暂无评论

HandRT: Simultaneous Hand Shape and Appearance Reconstruction with Pose Tracking from Monocular RGB-D Video

引用

IEEE Transactions on Pattern Analysis and Machine Intelligence 2025年 PP卷 PP页

作者： Kalshetti, Pratik Chaudhuri, Parag Indian Institute of Technology Bombay Department of Computer Science and Engineering India

We propose a method to reconstruct a personalized hand avatar, representing the user's hand shape and appearance, from a monocular RGB-D video of a hand performing unknown hand poses under unknown illumination. Our method, HandRT, jointly optimizes hand pose, shape, appearance, and lighting parameters using a physically-based shading model in a differentiable rendering framework incorporating Monte Carlo path tracing. HandRT extends our previous work, Intrinsic Hand Avatar, by relaxing the assumption of a known coarse hand pose and utilizing depth data in the optimization. Specifically, we introduce an articulated registration energy based on iterative closest point over the depth point cloud that enables reconstruction from unknown hand poses via tracking across frames. Thus, we can reconstruct the avatar from arbitrary poses with high accuracy without relying on an off-the-shelf 2D joint detector at each frame. Further, HandRT is capable of precisely tracking the reconstructed avatar from either RGB or RGB-D input. Our evaluation demonstrates that our method outperforms existing hand avatar reconstruction methods on all commonly used metrics while producing significantly accurate mesh compared to state-of-the-art hand mesh recovery methods by a large margin on public and our captured datasets. To facilitate future research, we release our code at https://***/pmkalshetti/handrt © 1979-2012 IEEE.

关键词： Mesh generation

来源：评论

学校读者我要写书评

暂无评论

Real-Time Video Enhancement with Spatio-Temporal Attention via Deep Convolutional Neural Networks 3rd

Real-Time Video Enhancement with Spatio-Temporal Attention v...

引用

3rd International Conference on Intelligent Systems and Sustainable Computing, ICISSC 2023

作者： Maria, Anto Bennet Balasubramanian, Swaminathan Vijayan, Rajmohan Sonthi, Vijaya Krishna Ramakrishnan, Jayapratha Manogaran, Umamaheswari Sengan, Sudhakar Department of Electronics and Communication Engineering Vel Tech Rangarajan Dr. Sagunthala R&D Institute of Science and Technology Tamil Nadu Chennai600062 India Karnataka Bengaluru560069 India Tamil Nadu Chennai602107 India Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation Andhra Pradesh Vaddeswaram522502 India Department of Information Technology V.S.B. College of Engineering Technical Campus Tamil Nadu Coimbatore642109 India School of Computer Science and Engineering Vellore Institute of Technology Tamil Nadu Vellore632014 India Department of Computer Science and Engineering PSN College of Engineering and Technology Tamil Nadu Tirunelveli627451 India

ISBN: (纸本)9789819783540

The realm of video content has witnessed exponential growth, and alongside it emerges the challenge of enhancing video quality efficiently. Traditional techniques, although valuable, often fall short of addressing the intricate details and temporal consistencies required for high-quality video upscaling. Moreover, real-time processing demands methods ensuring accuracy and swift computational speeds. Addressing these challenges, this paper introduces the STA-DCNN. This innovative architecture is rooted in combining spatial and temporal attention mechanisms to produce superior video super-resolution. The Spatio-Temporal Attention Module (STAM) acts as a cornerstone, singling out essential features across frames, while the Deep Convolutional Backbone (DCB) intensifies these frames, capitalizing on attention-informed features. Furthermore, the Temporal Memory Unit (TMU) preserves crucial aspects from preceding frames, guaranteeing a seamless video rendition. Empirical tests confirm that STA-DCNN notably surpasses its counterparts, offering an ideal balance of quality and real-time processing. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Spatio-temporal data

来源：评论

学校读者我要写书评

暂无评论

YOLO and BiLSTM Fusion for Smart Motorcycle Ignition

YOLO and BiLSTM Fusion for Smart Motorcycle Ignition

引用

International Conference on Inventive Computation Technologies (ICICT)

作者： E. Daniel L.H. Akilan Jainish G R P. Alwin Infant Daniel Elliot S Department of Computer science & Engineering Panimalar Engineering College Chennai City Campus Chennai India Department of Computer science & Engineering Loyola ICAM College of Engineering and Technology Chennai India Department of Computer Science & Engineering Panimalar Engineering College Chennai India

ISBN: (数字)9798331512248

ISBN: (纸本)9798331512255

Motorcycle accidents remain a leading cause of severe injuries and fatalities worldwide, underscoring the critical need for effective helmet enforcement mechanisms. This paper presents a novel and intelligent helmet recognition system designed to enhance road safety through accurate, real-time detection and compliance verification. Traditional systems based on Single Shot detection (SSD) and Convolutional neural networks(CNNs) often suffer from limitations in detection accuracy and latency, making them less viable for dynamic realworld applications. To overcome these challenges, the proposed system integrates the YOLO (You Only Look Once) architecture for fast and precise object detection with a Bidirectional Long Short-Term Memory (BiLSTM) network to capture and classify temporal features, enhancing decision-making over time. The Raspberry Pi receives the output of the helmet detection and uses it to control the motor cycle's ignition and trigger the relay. A key innovation of this system is a 60-second continuous verification module that ensures the motorcycle ignition activates only of a helmet is consistently detected, thereby minimizing false positives and promoting genuine compliance. Extensive experimental evaluation demonstrates substantial improvements in accuracy, processing speed and robustness across varying conditions, outperforming conventional approached. The proposed solution offers a practical, scalable, and efficient framework for integration into intelligent transportation system, with the ultimate goal of reducing head injuries and saving lives through proactive safety enforcement.

关键词： YOLO Head Accuracy Motorcycles Bidirectional long short term memory Ignition Real-time systems Safety Convolutional neural networks Injuries

来源：评论

学校读者我要写书评

暂无评论

Multi-Agent Path Finding Using Conflict-Based Search and Structural-Semantic Topometric Maps

arXiv

引用

arXiv 2025年

作者： Fredriksson, Scott Bai, Yifan Saradagi, Akshit Nikolakopoulos, George Robotics and AI group Department of Computer Science Electrical and Space Engineering Luleå University of Technology Sweden

As industries increasingly adopt large robotic fleets, there is a pressing need for computationally efficient, practical, and optimal conflict-free path planning for multiple robots. Conflict-Based Search (CBS) is a popular method for multi-agent path finding (MAPF) due to its completeness and optimality;however, it is often impractical for real-world applications, as it is computationally intensive to solve and relies on assumptions about agents and operating environments that are difficult to realize. This article proposes a solution to overcome computational challenges and practicality issues of CBS by utilizing structural-semantic topometric maps. Instead of running CBS over large grid-based maps, the proposed solution runs CBS over a sparse topometric map containing structural-semantic cells representing intersections, pathways, and dead ends. This approach significantly accelerates the MAPF process and reduces the number of conflict resolutions handled by CBS while operating in continuous time. In the proposed method, robots are assigned time ranges to move between topometric regions, departing from the traditional CBS assumption that a robot can move to any connected cell in a single time step. The approach is validated through real-world multi-robot path-finding experiments and benchmarking simulations. The results demonstrate that the proposed MAPF method can be applied to real-world non-holonomic robots and yields significant improvement in computational efficiency compared to traditional CBS methods while improving conflict detection and resolution in cases of corridor symmetries. Copyright © 2025, The Authors. All rights reserved.

关键词： Continuous time systems

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：