检索结果-内蒙古大学图书馆

SSRN 2024年

作者： Guo, Xiangxin Mu, Da School of Physics and Optoelectronic Engineering Changchun University of Science and Technology Jilin Changchun130022 China Photoelectric Measurement and Control Technology Institute Changchun University of Science and Technology Jilin Changchun130022 China

An automotive augmented reality head-up display (AR-HUD) system is affected by many complex phenomena, such as the influence of moving eyeballs, drivers’ heights, additional aberration correction required due to windshields, and variable structural constraints of vehicle model and other factors. Consequently, it is difficult to optimize the design of such a system. To mitigate this problem, we propose an automatic design method for the automotive AR-HUD imaging system, which is suitable for two freeform surfaces and any type of windshield structure. Based on Malus Law, all the characteristic data points of the freeform surface in the full field of view and full aperture are calculated. Subsequently, the points are fitted to the free-form surface represented by polynomials. The initial structure of the free-form off-axis system with a good imaging quality is directly obtained. Finally, the optical path feedback iterative method is used to further optimize the data point cloud of each field of view to determine the best fitting coefficients, generate the required freeform mirror, and obtain the final structure of the AR-HUD optical system. The proposed AR-HUD designs for an eye-box of and a field of view of exhibit excellent optical performance, demonstrating the feasibility and superiority of the proposed design framework. © 2024, The Authors. All rights reserved.

关键词： Augmented reality

来源：评论

学校读者我要写书评

暂无评论

Advanced vortex beam interference system enhanced by CNN for micro-displacement detection

引用

Optics and Lasers in engineering 2025年 191卷

作者： Yuan, Tao Liu, Ji Wu, Jinhui Zhang, Yiman Wang, Yujin Jing, Ning School of Information and Communication Engineering North University of China Taiyuan030051 China State Key Laboratory of Dynamic Measurement Technology North University of China Taiyuan030051 China School of Instrument and Electronics North University of China Taiyuan030051 China

This paper introduces an innovative conjugate vortex beam interference system that significantly enhances micro-displacement measurement capabilities by incorporating a deep learning model for processing and analyzing interference patterns. The theoretical foundation of this system is based on the petal-like interference pattern graduated when two conjugate vortex beams interact, where the rotation of the fringe orientation exhibits a strict linear correlation with displacement. The system fully exploits the sensitivity of vortex beams to minute changes in optical paths, along with the conjugate interference effect, to capture precisely angular variations in the interference pattern. This results in a measurement accuracy better than 0.84 nm across the 0–500 nm range, with a mean absolute error (MAE) approximately 0.5 nm, while achieving sub-nanometer precision across an extensive measurement range. By innovatively applying deep learning to analyze vortex beam interference data, the system effectively utilizes the unique properties of vortex beams alongside the powerful feature learning capabilities of convolutional neural networks (CNNs), enhancing robustness against fringe distortion and noise, thereby improving the accuracy of displacement detection. This method provides an accurate, efficient, and non-contact approach to displacement measurement, showcasing distinct advantages in the precision and stability of micro-displacement assessments. © 2025

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Automatic Design of Multi-Light Field Ar-Hud Based on Fresnel Ladder Freeform Mirror

SSRN

引用

SSRN 2024年

Conventional augmented reality head-up display (AR-HUD) suffers from various issues, such as large volume, single-image layer single-focal surface display, difficult optimization design of special face shape, and low definition under wide frame. To overcome these limitations, we propose an AR-HUD imaging system design method with Fresnel stepped free-form mirror, which can be adapted to any type of given windshield. First, the Malus' law is used to solve the discrete point cloud in the initial architecture. Second, the stepped Fresnel region is constructed by the field of view (FOV) extension method to directly obtain a good initial structure. Last, the optical path feedback iterative method is used to further optimize the data point cloud in each FOV and determine the best fitting coefficients. Using this method, two kinds of Fresnel AR-HUD structure layout, i.e., horizontal and vertical, are designed. Simulation results show that the proposed design framework can effectively reduce the volume by 20%, realize multi-layer distribution of image information display at different distances, integrate effectively with the AR, and improve image quality synchronization, which prove its feasibility and superiority. © 2024, The Authors. All rights reserved.

关键词： Ladders

来源：评论

学校读者我要写书评

暂无评论

A Multi-Scale SAR-Optical Image Matching Method Using Structure-Enhanced Convolutional Layer and Transformer-CNN Model

A Multi-Scale SAR-Optical Image Matching Method Using Struct...

引用

IEEE International Symposium on Geoscience and Remote Sensing (IGARSS)

作者： Yijun Liu Jie Long Qian Liu Chang Liu Qingsong Wang School of Electronics and Communication Engineering Shenzhen Campus of Sun Yatsen University Shenzhen China Beijing Institute of Control and Electronic Technology Beijing China

ISBN: (数字)9798350360325

ISBN: (纸本)9798350360332

Due to the geometric and nonlinear radiometric differences (NRD) between SAR and optical images, coupled with the impact of speckle noise, achieving robust and high-precision matching between them remains challenging. This paper introduces a novel multi-scale SAR-optical image matching method, which employs an improved multi-scale matching strategy. Specifically, we first establish an image pyramid and detect geometric keypoints of the images, subsequently segmenting the images into sub-blocks centered around these keypoints. These sub-blocks are then fed into the proposed siamese feature description network, comprising of a structure-enhanced convolutional layer and a Transformer-CNN model, for feature descriptors extraction, followed by feature matching. Finally, the matching point pairs are obtained by combining the matching relationships between image pairs of different scales. Experimental results show that our proposed algorithm achieves better matching performance compared to the existing state-of-the-art algorithms.

关键词： Image matching Speckle Optical fiber networks Feature extraction Transformers Optical imaging Radiometry

来源：评论

学校读者我要写书评

暂无评论

Indication Detection of Electric Energy Meter Based on YOLOv5

Indication Detection of Electric Energy Meter Based on YOLOv...

引用

作者： Bao-hua, Yang Kang-ming, Huang Yu, Sun Hong-jun, Guo School of Information Ensgineering Changzhou Vocational Institute of Mechatronic Technology Changzhou213164 China Department of Computer Engineering Asian University Suwon443-749 Korea Republic of Jiangsu Xinhuayu Measurement and Control Technology Co. Ltd Changzhou213164 China Jiangsu Xiangguo Automation Technology Co. Ltd Changzhou213164 China

With the continuous popularization of instruments, the inspection and testing of instruments have received more and more attention. This method is mainly divided into two links: target detection and information recognition. First, the instrument is located by OpenCV findcircuits function, and the instrument area is extracted. Then, the relevant methods of digital image processing are used for preprocessing, extracting the number information of the instrument panel, and establishing the data set. By using OpenCV to realize image edge detection, image segmentation and target area recognition, and then the deep learning methods are used are YOLOv3, YOLOv4 and YOLOv5 algorithms to realize the recognition of indications, the results show that the recall, precision, AP and F1 indicators of YOLOv5 were 95.26% and 95.89%, respectively, 96.29% and 95.77%, the above results of YOLOv5 are significantly improved compared with YOLOv3 and YOLOv4. The use of bottleneckcsp1 and bottleneckcsp2 in the neck can improve the network speed while ensuring accuracy, which shows that YOLOv5 is a very suitable real-time test method for instrument readings. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Reinforcement Learning-Based Fault-Tolerant control of Uncertain Strict-Feedback Nonlinear Systems With Intermittent Actuator Faults

引用

IEEE Transactions on Neural Networks and Learning Systems 2025年第6期36卷 10464-10478页

作者： Qinmin Yang Huaying Li Zhengwei Ruan Bo Fan Shuzhi Sam Ge State Key Laboratory of Industrial Control Technology College of Control Science and Engineering Zhejiang University Hangzhou China Hangzhou Huawei Communication Technology Company Ltd. Hangzhou China School of Automation Science and Engineering Xi’an Jiaotong University Xi’an China Department of Electrical and Computer Engineering Institute for Functional Intelligent Materials National University of Singapore Queenstown Singapore

In this work, a novel reinforcement learning-based adaptive fault-tolerant control (FTC) scheme with actuator redundancy is presented for a nonlinear strict-feedback system with nonlinear dynamics and uncertainties. A learning-based switching function technique is established to steer different groups of actuators automatically and successively to mitigate the impact of faulty actuators by observing a switching performance index. The optimal tracking control problem (OTCP) of strict-feedback nonlinear systems is transformed into an equivalent optimal regulation problem of each affine subsystem via adaptive feedforward controllers. Subsequently, the designed objective functions associated with Hamilton–Jacobi–Bellman (HJB) estimate errors caused by neural network (NN) approximations can be minimized by the reinforcement learning algorithm without value or policy iterations. It is proved that the tracking objective can be achieved and all signals in the closed-loop system can be guaranteed to be bounded, as long as the minimum time interval between two successive failures is bounded. Theoretical results are verified by simulations.

关键词： Actuators Switches Fault tolerant systems control systems Optimal control Redundancy Nonlinear dynamical systems Adaptive control Regulation Industrial control

来源：评论

学校读者我要写书评

暂无评论

Analysis of Visual Effects Configuration Scheme for Small Cultural Complex 2

Analysis of Visual Effects Configuration Scheme for Small Cu...

引用

2nd IEEE International Conference on Information technology, Big Data and Artificial Intelligence, ICIBA 2021

作者： Sun, Chang Su, Zhibin Liu, Wei Ren, Hui School of Information and Communication Engineering Communication University of China Beijing China Key Laboratory of Acoustic Visual Technology and Intelligent Control Ministry of Culture and Tourism School of Information and Communication Engineering Beijing China

ISBN: (纸本)9781665428767

Stage lighting design is an important step in the visual effect configuration scheme of small cultural complex. This paper mainly discusses the lighting configuration of small space with multi-functional art services under different modes such as performance, conference, exhibition, reading and folk customs. Focusing on the analysis of lighting layout methods in drama and artistic performances, this paper takes the drama performing environment as the representative. Through the corresponding subjective evaluation experiments, research and data analysis are carried out on the lamps and lighting methods adopted in the performance. Furthermore, the optimal configuration effect of front light and top light under given constraints is selected. The research provides an effective reference for the lighting configuration of small cultural complexes, which has certain application value and practical significance. © 2021 IEEE.

关键词： Lighting

来源：评论

学校读者我要写书评

暂无评论

Omnidirectional Vector measurement Method for Accurate Laser Positioning System

Omnidirectional Vector Measurement Method for Accurate Laser...

引用

IEEE International Conference of Intelligent Robotic and control engineering (IRCE)

作者： Xiao Pan Fei Liu Qing Liu Zhen Wang Jiapei Li Pengfei Cheng Xi'an Aerocomm Measurement and Control Technology Co. Ltd. Xi'an China School of Materials Science and Engineering Xi'an University of Science and Technology Xi'an China School of Automation and Information Engineering Xi'an University of Technology Xi'an China School of Mechanical and Precision Instrument Engineering Xi'an University of Technology Xi'an China

ISBN: (数字)9798350352399

ISBN: (纸本)9798350352405

The accurate laser positioning system (ALPS) is an emerging distributed large-scale laser positioning system, which has shown broad prospects in industrial production, navigation, and other fields. Addressing the issue of reduced measurement accuracy in complex large-space measurement scenarios due to laser signal obstruction encountered by traditional planar and spherical receiver units, this paper analyzes the measurement principle of ALPS. Based on omnidirectional photoelectric array units, a comprehensive vector measurement model is established, and a position and attitude calculation method for omnidirectional vector measurement is studied. Employing the Monte Carlo method based on statistical principles, the feasibility of the position and attitude calculation method is verified through simulation. Additionally, field experiments are conducted in conjunction with a laser tracking system to evaluate the absolute accuracy of the proposed model under point and attitude changes. The results indicate that the point accuracy exceeds 0.24mm with a standard deviation of less than 0.026mm. When attitude changes occur, the horizontal and pitch directions are superior to 0.28mm and 0.47mm, respectively, with standard deviations of less than 0.030mm and 0.032mm, meeting the system's sub-millimeter-level requirements and expanding the application scope of the ALPS system.

关键词： Accuracy Monte Carlo methods measurement by laser beam Receivers Position measurement Laser stability Laser modes Vectors Laser fusion Standards

来源：评论

学校读者我要写书评

暂无评论

Study on heat transfer failure time effect energy of lithium heat pipe 3

Study on heat transfer failure time effect energy of lithium...

引用

2023 3rd International Conference on Advanced Materials and Mechatronics, ICAMM 2023

作者： Zhang, Yitong Zhou, Tao Liu, Wenbin Wei, Dong Mao, Shang Xue, Chunhui Lu, Huaichang Yao, Yao Department of Nuclear Science and Technology School of Energy and Environment Southeast University Nanjing210096 China Nuclear Thermal Safety and Standardization Team China National Engineering Research Center for Safe Operation and Intelligent Measurement and Control of Large Power Generation Equipment Nanjing210096 China

Studying the efficiency of lithium heat pipe above 1000K in case of problems is of considerable value for mastering and controlling its safety. We use COMSOL software to model and calculate a single element above 1000 K, and calculate and analyze its various parameter changes under fault conditions. According to different situations that should be classified and discussed, the cold end fluid temperature and corresponding heat transfer coefficient in the lower temperature area, as well as the temperature situation at the other end, should be separately considered. Based on the calculated results, it is discussed that they all undergo linear and proportional changes with the target parameters, and the fluid situation on the lower temperature side has the smallest impact on these target results. The maximum temperature of the module will reach 1900K, and materials that can withstand 1900K need to be selected during design. © Published under licence by IOP Publishing Ltd.

关键词： Heat pipes

来源：评论

学校读者我要写书评

暂无评论

Informer neural network-based prediction of degradation trajectory of energy meter clock battery life

Informer neural network-based prediction of degradation traj...

引用

14th International Conference on Quality, Reliability, Risk, Maintenance, and Safety engineering (QR2MSE 2024)

作者： Lingyu Zhang Yongqi Chen Xinkun Qin Wanlin He Jia Qi Yongquan Sun School of Computer Science and Technology Harbin University of Science and Technology Harbin People's Republic of China School of Measurement and Communication Engineering Harbin University of Science and Technology Harbin People's Republic of China School of Electrical and Electronic Engineering Harbin University of Science and Technology Harbin People's Republic of China

ISBN: (数字)9781837241958

Accurate prediction of the life degradation trajectory of the clock battery, a critical component of the power meter, is essential for assessing the reliability of the device. Addressing the computational inefficiency and limited generalization capabilitie s of traditional empirical recession models, this paper introduces a prediction model based on the Informer neural network. The battery degradation data are initially segmented and reassembled using a sliding window strategy. The encoder-decoder structure is then employed to achieve multilevel feature extraction from the input sequences. By setting various prediction starting points, the model demonstrates high sensitivity in predicting battery degradation under temperature conditions identical to those in the training set. Additionally, the model is capable of predicting the degradation trajectories of batteries under different temperature conditions, such as 60℃.

关键词：

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：