检索结果-内蒙古大学图书馆

Onboard Person Retrieval System With Model Compression: A Case Study on Nvidia Jetson Orin AGX

IEEE ACCESS 2025年 13卷 8257-8269页

作者： Chaudhari, Jay N. Galiyawala, Hiren Sharma, Paawan Shukla, Pancham Raval, Mehul S. Ahmedabad Univ Sch Engn & Appl Sci Ahmadabad 380009 India RyDOT Infotech Pvt Ltd Ahmadabad 380027 India Pandit Deendayal Energy Univ Sch Technol Gandhinagar 382007 India Imperial Coll London London SW7 2AZ England

A person retrieval system (PRS) in video surveillance identifies an individual based on descriptive attributes, a task that employs several computationally intensive deep learning models. We implement and analyse a PRS for pre-recorded videos on a graphics processing unit (GPU) and Nvidia Jetson Orin AGX. This paper presents a new Person Attribute Recognition (PAR) architecture, CorPAR, using three backbone networks, ConvNext, ResNet-50, and EfficientNet-B0. It enhances the F1-score by 4.1% with ConvNeXT-Base, 1.63% with the ResNet, and by 8.07% with EfficientNet-B0, surpassing the performance of the state-of-the-art Weighted-PAR method. The proposed method uses model compression techniques like quantisation and pruning with L1 regularisation to assess their impact on person retrieval. The study reveals that the PRS utilising EfficientNet-B0, with 32-bit quantisation, achieves the best performance, delivering a throughput of 22 frames per second and a True Positive Rate of 71% on Nvidia Jetson Orin AGX matching the performance of a model implemented using GPU.

关键词： Clothing image edge detection Surveillance Quantization (signal) Real-time systems performance evaluation image color analysis Graphics processing units Videos Computational modeling Edge device model compression person attribute recognition person retrieval pruning quantization surveillance

来源：评论

学校读者我要写书评

暂无评论

Enhancing Object Detection in Assistive Technology for the Visually Impaired: A DETR-Based Approach

引用

IEEE ACCESS 2025年 13卷 71647-71661页

作者： Ikram, Sunnia Sarwar Bajwa, Imran Gyawali, Sujan Ikram, Amna Alsubaie, Najah Islamia Univ Bahawalpur Dept Comp Sci Bahawalpur 63100 Pakistan Lamar Univ Dept Comp Sci Beaumont TX 77705 USA Govt Sadiq Coll Women Univ Dept Comp Sci Bahawalpur 63100 Pakistan Princess Nourah bint Abdulrahman Univ PNU Coll Comp & Informat Sci Dept Comp Sci POB 84428 Riyadh 11671 Saudi Arabia

This paper presents a real-time obstacle detection and recognition system designed to enhance navigation for visually impaired individuals through assistive technology. The system integrates a mobile application equipped with a mini camera for real-time image capture and employs advanced deep learning techniques for object detection and classification. A comparative evaluation of YOLOv8, Faster R-CNN and DETR (Detection Transformer) is conducted based on precision, Recall, F1-score, confidence score and processing efficiency. DETR demonstrates superior performance, achieving a 99% confidence score, 98% precision and a processing speed of 40ms per frame. While faster R-CNN and YOLOv8 provide competitive results, they offer a trade-off between accuracy and computational efficiency. The system follows a structured a structured workflow, including real-time acquisition, preprocessing, innovative data augmentation and optimization for edge devices using TensorFlow Lite for efficiency deployment. It classifies 80 types of obstacles, such as pedestrians, vehicles and traffic signal and provides immediate audio feedback to ensure safe navigation. The model trained over 20 epochs achieves an accuracy of 98% in the final epoch. This study introduces a scalable and practical solution integrating IoT and real-time image processing, empowering visually impaired users with enhanced mobility and safety.

关键词： Sensors Navigation Accuracy Real-time systems Cameras Mobile applications Sensor systems Safety Detection algorithms Computational modeling Computer vision detection transformer (DETR) region-based convolutional neural network (R-CNN) IoT-based object detection performance evaluation of deep learning models real-time object recognition You Only Look Once (YOLOv8)

来源：评论

学校读者我要写书评

暂无评论

modeling of Hybrid Fiber Optic and Free-Space Optical Communication systems 31

Modeling of Hybrid Fiber Optic and Free-Space Optical Commun...

引用

31st IEEE Conference on signal processing and Communications Applications (SIU)

作者： Evran, Serap Kilinc Unverdi, N. Özlem Yildiz Tekn Univ Elekt & Haberlesme Muhendisligi Istanbul Turkiye

ISBN: (纸本)9798350343557

Free-space optical (FSO) communications systems have many advantages, such as large bandwidth, high data transmission rate, low cost, and easy installation process. In this study, a hybrid system incorporating both wired optical communication and free-space optical communication was designed, and its performance was analyzed. OptiSystem 20.0 simulation program was utilized in designing and performance evaluation of the hybrid communication system.

关键词： Communication technologies fiber optic communication free-space optical communications laser amplifier

来源：评论

学校读者我要写书评

暂无评论

Analysis of the Influence of Noise in Reference signal on the performance of Non-Cooperative Passive Detection 17

Analysis of the Influence of Noise in Reference Signal on th...

引用

17th International Congress on image and signal processing, BioMedical Engineering and Informatics, CISP-BMEI 2024

作者： Ying, Tao Zhang, Ruoyu Zhou, Cheng Hou, Xiaoyang College of Electronic Engineering Naval University of Engineering Wuhan430033 China

ISBN: (纸本)9798331507398

In non-cooperative passive detection systems, reference signals are inevitably interfered by noise and will affect the detection performance of the system. This paper carries out in-depth analysis and research on this issue. First, the digital radio mondiale signal is used as the radiation source, the signal model and ambiguity function performance of digital radio mondiale signal are analyzed. Then, the influence of noise on the detection performance of the system is analyzed by theoretical modeling and formula deduction, based on the model of reference signal and surveillance signal. Finally, the simulation results verify the relevant conclusions of the theoretical derivation. © 2024 IEEE.

关键词： signal analysis

来源：评论

学校读者我要写书评

暂无评论

simulation and Fabrication Feasibility of Two-Channel All-Optical Modulator Exploiting InAs/AlAs Colloidal Quantum Dots

引用

IEEE ACCESS 2024年 12卷 126159-126175页

作者： Faraji, Mohammad Matloub, Samiye Dortaj, Hannaneh Univ Tabriz Fac Elect & Comp Engn Quantum Photon Res Lab QPRL Tabriz *** Iran

In the context of advancing photonics technologies, there is an increasing demand for broadband and ultrafast optical modulators to support next-generation telecommunication networks and ultrafast quantum information processing. Existing modulators are limited by their inability to efficiently handle multiple channels, leading to challenges in high-speed optical communication systems. We introduce a novel two-channel quantum dot all-optical modulator (QD-AOM) designed to simultaneously modulate telecommunication wavelengths of 1.32 mu m and 1.55 mu m by utilizing the interaction between two pump signals (683 nm and 866 nm) and two probe signals (1.32 mu m and 1.55 mu m). This innovative device leverages InAs/AlAs core/shell colloidal quantum dots, offering a scalable, cost-effective, and efficient solution for next-generation optical communication networks and ultrafast quantum information processing. Theoretical modeling was performed by solving the 3D Schrodinger equation, coupled rate and propagation equations to simulate the device's performance. The model accounts for carrier transfer processes and the interaction between channels through Fluorescence Resonance Energy Transfer (FRET). simulations demonstrated that the QD-AOM achieved modulation depths of 65% and 62% for channels corresponding to 1.55 mu m and 1.32 mu m, respectively, with bandwidths of 26.3 GHz and 19.7 GHz. The device maintained high stability and efficiency at room temperature, with minimal variation in modulation depth across different temperatures, indicating its robustness for practical applications. Its broadband response, high modulation depth, and robust performance make it an ideal candidate for integration into future high-speed communication systems, signal processing, and other photonics-based applications.

关键词： Modulation Probes Mathematical models Optical modulation Quantum dots Photonics performance evaluation All-optical modulator colloidal quantum dots high modulation depth

来源：评论

学校读者我要写书评

暂无评论

A Novel Frequency Measurement Methodology for Clock Synchronization in SPICE-Based Simulators

引用

IEEE ACCESS 2023年 11卷 117030-117039页

作者： Guran, Ionut-Constantin Florescu, Adriana Perisoara, Lucian Andrei Univ Politehn Bucuresti Fac Elect Telecommun & Informat Technol Dept Appl Elect & Informat Engn Bucharest 060042 Romania

Nowadays, there is an increasing demand for complex behavioral models on the market, as simulation has become the main verification approach used for both analog and digital functionalities. One of the most popular mixed-signal simulators nowadays is simulation Program for Integrated Circuits Emphasis (SPICE). Over the last decade, the digital part in circuits has become predominant, but this poses big challenges when developing SPICE simulation models, as the circuits are synchronized upon internal or external clock signals, where the frequency accuracy is mandatory. The existing SPICE based simulators have built-in options to measure the signal frequency and duty cycle, but only after the simulation has completed, and in this way the frequency information can not be used during the simulation to synchronize the clock signals. There is a lot of research on frequency measurement techniques in the literature, but this topic has not been translated and covered in SPICE until now. Implementing the physical frequency measurement circuits proposed previously in the literature would increase the models' complexity and the simulation time, without benefitting the final user. For this reason, in this article, a novel frequency measurement methodology for clock synchronization in SPICE based simulators is proposed, which aims to provide accurate measured frequency and duty cycle values for the digital clock signals with a fast simulation time. The performance evaluation of this methodology is done using OrCAD Capture, which is the most popular SPICE-based simulation environment.

关键词： Frequency measurement Clocks SPICE Synchronization Integrated circuit modeling Integrated circuits image edge detection Digital signal processing simulation Clock synchronization digital signals frequency measurement SPICE simulation models

来源：评论

学校读者我要写书评

暂无评论

Theoretical and Experimental Investigation of Using Multidegree of Freedom Electrostatically Actuated Microstructures in Performing Classification Problems

引用

IEEE SENSORS JOURNAL 2023年第11期23卷 12049-12057页

作者： Nikfarjam, Hamed Megdadi, Mohammad Okour, Mohammad Pourkamali, Siavash Alsaleem, Fadi Univ Texas Dallas Dept Elect Engn & Comp Sci Richardson TX 75080 USA Univ Nebraska Lincoln Dept Mech Engn Lincoln NE 68588 USA Univ Nebraska Lincoln Dept Architectural Engn Lincoln NE 68588 USA

This article demonstrates the use of micro-electromechanical systems (MEMS) to perform an efficient classification problem. Classification using digital computers is inefficient due to its power requirement and the separation between memory and processing units. As a proof of concept, the basic signal classification has been performed using a network of only three mechanically coupled devices. The classification has occurred at the hardware level rather than using a digital computer. The classification problem involves distinguishing between a ramp signal and a step input signal. In this demonstration, a three-degree-of-freedom model was built for the proposed network in MATLAB. The detailed simulation results were extracted to validate the network operation and sensitivity to operating conditions. Next, the device network hardware was fabricated and tested to perform the required classification problem. Good agreement between simulation and experiment is demonstrated.

关键词： Micromechanical devices Voltage Mathematical models performance evaluation Computational modeling Hardware Electrostatics Mechanical computing mechanical coupling micro-electromechanical systems (MEMS) neuromorphic computing parallel-plate actuators pull-in signal classification

来源：评论

学校读者我要写书评

暂无评论

Battery-less RF-powered circuits for non-contact voltage monitoring of electric systems: Circuit modeling and SPICE analysis

引用

INTERNATIONAL JOURNAL OF CIRCUIT THEORY AND APPLICATIONS 2025年第2期53卷 840-860页

作者： Colella, Riccardo Catarinucci, Luca Grassi, Giuseppe Natl Res Council Italy Branch Lecce Lecce Italy Univ Salento Dept Innovat Engn Lecce Italy CNIT Lecce Italy Univ Salento Dept Innovat Engn Lecce Italy

Monitoring electric power systems is essential for avoiding disruptions, improving efficiency, and detecting faults. Traditional methods involve interrupting the power lines and pose safety risks. On the contrary, non-contact power meters allow for safe power measurement without touching the wires but designing them presents challenges related to battery life and sustainability. This paper introduces a novel battery-less RF-powered wireless circuit that overcomes these limitations and optimizes the performance of the sensing of the voltage. This circuit includes an efficient RF-DC converter, a designed non-contact flexible and curved AC voltage sensor, a signal conditioning circuit, a digital processing unit, and a passive UHF Radio-Frequency Identification interface for wireless communication. It harnesses electromagnetic energy from the radio front end to power the entire system. The circuit has been properly conceptualized, developed, and analyzed through electrical modeling and SPICE simulations. Transient and steady-state analyses have been carefully conducted on individual circuit blocks to analyze their electric characteristics. Finally, a proof-of-concept circuit has been realized and experimentally verified to validate the simulation outcomes. With a bandwidth of 3 kHz, an RF-DC conversion sensitivity of -16 dBm, a current consumption of only 2.2 mA, and an efficient power management procedure, this energy-autonomous, non-contact metering circuit offers very high RF and sensing sensitivities and ultra-low power consumption, making it ideal for monitoring AC electric systems, particularly in inaccessible environments where battery-based solutions are impractical. Architecture of the novel battery-less RF-powered wireless voltage meter optimized to perform accurate voltage sensing from AC insulated wires of electric power lines and electric distribution systems. image

关键词： circuit model electric system non-contact voltage sensing power line RFID SPICE model transient solver

来源：评论

学校读者我要写书评

暂无评论

Fractional Order Convex Linear Prediction for signal Modelling

引用

CIRCUITS systems AND signal processing 2024年第9期43卷 5577-5590页

作者： Alaviyan Shahri, Seyed Mostafa Abutalebi, Hamid Reza Yazd Univ Dept Elect Engn Univ Blvd Yazd *** Iran

This paper introduces and analyzes three innovative Linear Prediction (LP) models, consisting of two Fractional Order (FO) LP models and an integer LP model. The closed-form expressions for the proposed methods-namely, Two Parameter FO Linear Prediction (TP-FLP), FO Convex Linear Prediction (FCLP), and integer Convex Linear Prediction (CLP)-are derived for derivative order 0 <=alpha <= 2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ 0 \le \alpha \le 2 $$\end{document}. These derivations contribute to a more profound understanding of underlying mathematical principles of the proposed predictor. The foundation of our proposed methodologies is based on the presentation of the current signal sample as a linear combination of the FO derivative of two past samples, coupled with the utilization of the convex combination technique. An extensive set of experiments is conducted to compare the effectiveness of the proposed models with baseline methods involving two traditional LP models: First- and Second-order LP, along with the One Parameter FO Linear Prediction (OP-FLP) model. The performance evaluation encompasses a diverse range of signal types, including sinusoidal waves, damped sine waves, electroencephalogram (EEG), and speech signals. The simulation results demonstrate significant advancements achieved by the proposed models. Notably, the proposed FCLP exhibits a remarkable prediction gain of 24.74 dB, outperforming baseline methods such as OP-FLP, First- and Second-order LP models, that achieve prediction gains of 23.17, 23.15, and 10.14 dB, respectively.

关键词： Fractional calculus Convex combinaton signal modeling Linear prediction

来源：评论

学校读者我要写书评

暂无评论

modeling and performance evaluation of TomoSAR System Based on Reflected signal of Beidou Navigation Satellite 14th

Modeling and Performance Evaluation of TomoSAR System Based ...

引用

14th China Satellite Navigation Conference, CSNC 2024

作者： Wang, Chenghao Liu, Feifeng Hu, Cheng Wang, Zhanze Xu, Zhixiang Radar Research Laboratory School of Information and Electronics Beijing Institute of Technology Beijing100081 China Ministry of Education Beijing100081 China

ISBN: (纸本)9789819969272

TomoSAR improves the information dimension of radar image from 2D to 3D, and is widely used in elevation modeling, deformation inversion of infrastructure such as high-rise buildings, dams and mining areas, and forest vegetation biomass estimation. However, the revisiting time of traditional spaceborne TomoSAR is more than 10 days, it is difficult to monitor and forecast the sudden deformation in real time. Furthermore, most of the low-orbit satellites are deployed in the north-south direction, so the deformation measurement accuracy in the north-south direction is relatively poor. The orbit baselines in height direction of low orbit satellites are small, so the image quantity and cost of high resolution tomography processing are huge. Therefore, low orbit TomoSAR is difficult to achieve high frequency, high precision and low-cost three-dimensional deformation measurement. Although GEO SAR and MEO SAR systems have the advantages of shorter revisit period, longer observation time and larger beam coverage area theoretically, there are no satellites in orbit at present. GNSS-TomoSAR, based on the reflected signals of navigation satellites, uses the in-orbit navigation satellite as the irradiation source, and a stationary receiver is placed on the ground to receive the echo signal reflected from the scene for synthetic aperture radar tomographic processing, which can achieve high frequency, high precision and low-cost three-dimensional deformation monitoring of the scene. GNSS-TomoSAR is a relatively novel signal processing concept, which is rarely studied at home and abroad. In this paper, the signal processing model of GNSS-TomoSAR is established, the performance boundary of GNSS-TomoSAR is discussed, and the actual track of Beidou IGSO satellite is used to conduct the simulation of height inversion of point target in GNSS-TomoSAR configuration. In above, the application prospect of GNSS-TomoSAR technology is preliminarily explored, lying a foundation for GNSS-TomoSAR

关键词： Synthetic aperture radar

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：