检索结果-内蒙古大学图书馆

Practical aberration correction using deep transfer learning with limited experimental data

Optics Express 2025年第6期33卷 14431-14444页

作者： Kok, Yong En Bentley, Alexander Parkes, Andrew J. Somekh, Michael G. Wright, Amanda J. Pound, Michael P. University of Nottingham School of Computer Science NottinghamNG8 1BB United Kingdom Optics and Photonics Group Department of Electrical and Electronic Engineering University of Nottingham NottinghamNG7 2RD United Kingdom Research Center for Humanoid Sensing Zhejiang Laboratory Hangzhou3111100 China

Adaptive optics is a technique for correcting aberrations and improving image quality. When adaptive optics was first used in microscopy, it was common to rely on iterative approaches to determine the aberrations present. It is advantageous to avoid iteration, and therefore there has been a shift to deep learning for aberration prediction. However, issues remain regarding the practicalities of machine learning for adaptive optics, an important one being the requirement for a large training dataset. Here, we explore transfer learning to overcome this need for data by pre-training a network on a large simulated dataset and fine-tuning it with reduced experimental data for application in an experimental setting. We demonstrate that the pre-trained network can make noticeable improvements with fine-tuning on just 24 experimental samples. To further enhance practicality, we significantly extend the range of aberrations present, predicting up to 25 Zernike modes with each coefficient ranging from −1 to 1, and perform a thorough analysis of the type and magnitude of phase-diversity required in the input data for a successful network. Our approach demonstrates substantial aberration reduction on experimental data for 10 Zernike modes, with an average 73% decrease in RMS wavefront error from 1.81 to 0.48 rad when correction is applied. This method achieves complete experimental image capture and aberration inference at rates comparable to the image acquisition time of a typical laser scanning microscope. Additionally, we consider the benefits of further improvements via an iterative step. As such, this work addresses some of the key practical hurdles that remain in the use of deep learning for aberration prediction and correction. © 2025 Optica Publishing Group (formerly OSA). All rights reserved.

关键词： Adaptive optics

来源：评论

学校读者我要写书评

暂无评论

Causes and Consequences of IPD Mismatch in XR Devices

Causes and Consequences of IPD Mismatch in XR Devices

引用

International Symposium, Seminar, and Exhibition, Display Week 2024

作者： Wilcox, Laurie M. Allison, Robert S. Department of Psychology York University TorontoM3J1P3 Canada Dept. of Electrical Engineering and Computer Science York University TorontoM3J1P3 Canada

Binocular 3D display systems, including virtual, augmented and mixed reality (collectively referred to as XR) devices support stereoscopic depth perception by presenting dichoptic views separately to the two eyes rendered according to their vantage points. A common human factors issue is a mismatch between the observer’s physical interpupillary distance (IPD) and 1) the separation between the device optics and 2) the rendered/captured content. This paper provides a theoretical review of the importance and consequences for user experience of accurately estimating IPD and incorporating it into HMD adjustments, calibration and rendering. © 2024, John Wiley and Sons Inc. All rights reserved.

关键词： Stereocenters

来源：评论

学校读者我要写书评

暂无评论

Lightweight secure image encryption: a tent map chaos theory approach

引用

Multimedia Tools and Applications 2025年 1-20页

作者： Odeh, Ammar Taleb, Anas Abu Alhajahjeh, Tareq Navarro, Francisco Ayesh, Aladdin Department of Computer Science Princess Sumaya University of Technology Amman1196 Jordan Faculty of Computing Engineering and Media De Montfort University LeicesterLE1 9BH United Kingdom Faculty of Computing University of Aberdeen King’s College AberdeenAB24 3 FX United Kingdom

In the digital era, securing image data against unauthorized access is a critical challenge, particularly in applications such as cloud storage, IoT security, and medical imaging. This study presents a lightweight and secure image encryption algorithm based on the Tent Map chaotic system, which enhances encryption strength while maintaining computational efficiency. The proposed approach effectively obfuscates image content, achieving a uniform histogram distribution indicative of robust encryption. Comparative analysis with existing methods demonstrates its superiority, particularly in security performance metrics, including entropy (7.996), NPCR (99.65%), and UACI (33.4724%). Additionally, image quality assessments using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity Index (SSIM) confirm that the algorithm preserves image integrity while ensuring strong encryption. The algorithm’s lightweight nature makes it suitable for large-scale applications, such as real-time encryption in cloud environments, IoT security, and multimedia protection. These results highlight the effectiveness of the proposed method in balancing security, efficiency, and image quality, making it a practical solution for modern image encryption challenges. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2025.

关键词： Signal to noise ratio

来源：评论

学校读者我要写书评

暂无评论

Microwave characterization of two Ba_(0.6)Sr_(0.4)TiO_(3) dielectric thin films with out-of-plane and in-plane electrode structures

引用

Journal of Advanced Ceramics 2023年第8期12卷 1521-1532页

作者： Hanchi Ruan Theo Graves Saunders Henry Giddens Hangfeng Zhang Achintha Avin Ihalage Jonas Florentin Kolb Matthew Blunt Sajad Haq Haixue Yan Yang Hao School of Electronic Engineering and Computer Science Queen Mary University of LondonLondon E14NSUK School of Engineering and Materials Science Queen Mary University of LondonLondon E14NSUK Department of Chemistry University College LondonLondon WC1H 0AJUK Advanced Services and Products QinetiQFarnborough GU140LXUK

Ferroelectric(FE)thin films have recently attracted renewed interest in research due to their great potential for designing novel tunable electromagnetic devices such as large intelligent surfaces(LISs).However,the mechanism of how a polar structure in the FE thin films contributes to desired tunable performance,especially within the microwave frequency range,which is the most widely used frequency range of electromagnetics,has not been illustrated *** this paper,we described several straightforward and cost-effective methods to fabricate and characterize Ba_(0.6)Sr_(0.4)TiO_(3)(BST)thin films at microwave *** prepared BST thin films here exhibit homogenous structures and great tunability(h)in a wide frequency and temperature range when the applied field is in the out-of-plane *** high tunability can be attributed to high concentration of polar *** response to the applied direct current(DC)field was directly visualized using a novel non-destructive near-field scanning microwave microscopy(NSMM)*** results have provided some intriguing insights into the application of the FE thin films for future programmable high-frequency devices and systems.

关键词： ferroelectric(FE) microwave tunability polarization thin film

来源：评论

学校读者我要写书评

暂无评论

Development of a Current-Position Control Strategy for Motion Systems Utilizing Nonlinear Reluctance Actuators 63

Development of a Current-Position Control Strategy for Motio...

引用

63rd IEEE Conference on Decision and Control, CDC 2024

作者： Al Saaideh, Mohammad Al-Rawashdeh, Yazan M. Alatawneh, Natheer Aljanaideh, Khaled Boker, Almuatazbellah Al Janaideh, Mohammad Memorial University St. John's Department of Electrical and Computer Engineering NLA1B 3X5 Canada University of Jordan Electrical Engineering Department at Al-Zaytoonah 130 Amman 11733 Jordan Memorial University St. John's Department of Mechanical and Mechatronics Engineering NLA1B 3X5 Canada Jordan University of Science and Technology Faculty of Engineering Department of Aeronautical Engineering Irbid22110 Jordan Virginia Tech Department of Electrical and Computer Engineering BlacksburgVA24060 United States University of Guelph School of Engineering GuelphON N1G 2W1 Canada

ISBN: (纸本)9798350316339

Reluctance actuators (RA) can replace the current Lorentz actuators in the next generation of positioning and scanning motion systems, such as the wafer stage in lithography machines. However, the nonlinear output force characteristic and the gap dependency of the RA are the main challenges in using the RA to drive motion systems. In this paper, we design a two-loop control approach for a reluctance actuator motion system (RAMS) to achieve tracking performance for a desired motion profile. First, the current control loop linearizes the RA under different conditions. Next, the position control loop is designed using a PID control based on an extended-high gain observer to achieve a desired motion profile considering unknown dynamics in the system. The simulation results show the efficiency of the proposed current control in linearizing the dynamic behavior under different desired forces and nominal air gaps and achieving a frequency response similar to the spring-mass-damper system. Moreover, the position control can achieve different long-stroke and short-stroke motion profiles with different amplitudes and ranges of motion. © 2024 IEEE.

关键词： Proportional control systems

来源：评论

学校读者我要写书评

暂无评论

Multi-Variate Time Series Prediction of Traffic and Users for Dynamic RRH-BBU Mapping in C-RAN

引用

IEEE Transactions on Mobile Computing 2025年

作者： Wu, Fan Wang, Shanshan Zhou, Jieyu Pan, Haoye Zhou, Conghao Yang, Wang Lyu, Feng Zhang, Yaoxue Central South University School of Computer Science and Engineering Changsha410083 China Rutgers University Department of Computer Science PiscatawayNJ United States University of Waterloo Department of Electrical and Computer Engineering WaterlooONN2L 3G1 Canada BNRist Tsinghua University Department of Computer Science and Technology Beijing100084 China Zhongguancun Laboratory Beijing100094 China

Cellular operators face significant challenges in cutting operating expenses while maintaining the quality of service (QoS) for users due to growing network traffic and dynamic user connections. These challenges are addressed by the cloud radio access network (C-RAN) architecture, which includes a centralized pool of baseband units (BBUs) and distributes them from remote radio heads (RRHs). The key to improving C-RAN performance is to dynamically allocate large-scale RRHs to different BBUs in real time. In this paper, we propose a user behavior-aware RRH-BBU mapping framework to improve the performance of large-scale C-RANs by predicting RRH traffic and users in advance. First, we propose a Multivariate RRH time series Prediction Model (MRPM) that captures the spatio-temporal patterns in the data to predict the traffic volume and the number of users of RRHs, which represents key indicators of RRH connection states. Second, we formulate the RRH-BBU mapping as a Markov decision process problem to optimize cost and QoS by considering BBU utilization, BBU energy consumption, RRH migration frequency, and BBU load balancing. Third, we propose a prediction-based RRH-BBU mapping scheme (PB-RBM) to find the optimal RRH-BBU mapping strategy by leveraging the prediction information of MRPM. In the PB-RBM algorithm, we employ an A3C algorithm to learn the mapping policy and group the RRHs based on a defined popularity metric to reduce the state and action space of the reinforcement learning algorithm. Finally, extensive experiments are conducted on a real-world dataset, and our algorithm is compared with several matching algorithms, such as ACKTR, heuristic, etc., to demonstrate its superiority, especially reducing 17.5% in RMSE compared to the best-performing baseline. © 2002-2012 IEEE.

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

A Method to Classify Emotions Based on BERT 4

A Method to Classify Emotions Based on BERT

引用

4th International Signal Processing, Communications and engineering Management Conference, ISPCEM 2024

作者： Chen, Zeli Lin, Jian Jiang, Jie Jing, Di Department of Computer Science Lafayette College EastonPA United States School of Information Technology and Artificial Intelligence Zhejiang University of Finance and Economics Zhejiang Hangzhou China Department of Mathematical and Statistical Sciences University of Alberta EdmontonABT6G 2G1 Canada School of Computer Science and Communication Engineering Jiangsu University Zhenjiang China

ISBN: (纸本)9798331528676

Nowadays, as science and technology continue to advance, addressing people's emotional issues is of particular significance, and how to accurately and efficiently recognize emotions has become increasingly prominent. We explores sentiment recognition and classification methods, including rule-based approaches, traditional machine learning models, deep learning models, and multi-modal sentiment analysis. Through comparative analysis, we will evaluate the effectiveness of these methods in handling social media texts. Additionally, we has developed a BERT-based two-stage model that first determines sentiment polarity (positive or negative) and then performs detailed classification of negative emotions. This two-stage model significantly improves accuracy and detail recognition compared to traditional one-stage models and other baseline methods. The results show that our approach not only enhances sentiment analysis accuracy but also provides deeper insights into the emotional subtleties of social media texts. This advancement holds significant practical implications for real-time sentiment monitoring and more effective crisis management strategies. ©2024 IEEE.

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

A Proposed Hybrid Recommender System for Tourism Industry in Iraq Using Evolutionary Apriori and K-means Algorithms

arXiv

引用

arXiv 2025年

作者： Hassan, Bryar A. Hassan, Alla A. Lu, Joan Ahmed, Aram M. Rashid, Tarik A. Computer Science and Engineering Department School of Science and Engineering University of Kurdistan Hewler Erbil Iraq Department of Computer Science College of Science Charmo University Sulaimani KRI Chamchamal46023 Iraq Department of Database Technology Computer Science Institute Sulaimani Polytechnic University Sulaimani Iraq Department of Computer Science School of Computing and Engineering University of Huddersfield Queensgate HuddersfieldHD1 3DH United Kingdom

The rapid proliferation of tourism data across sectors, including accommodations, cultural sites, and events, has made it increasingly challenging for travelers to identify relevant and personalized recommendations. While traditional recommender systems such as collaborative, content-based, and context-aware systems offer partial solutions, they often struggle with issues like data sparsity and overspecialization. This study proposes a novel hybrid recommender system that combines evolutionary Apriori and K-means clustering algorithms to improve recommendation accuracy and efficiency in the tourism domain. Designed specifically to address the diverse and dynamic tourism landscape in Iraq, the system provides personalized recommendations and clusters of tourist destinations tailored to user preferences and contextual information. To evaluate the system’s performance, experiments were conducted on an augmented dataset representative of Iraq’s tourism activity, comparing the proposed system with existing methods. Results indicate that the proposed hybrid system significantly reduces execution time by 27-56% and space consumption by 24-31%, while achieving consistently lower Root Mean Square Error (RMSE) and Mean Absolute Error (MAE) values, thereby enhancing prediction accuracy. This approach offers a scalable, context-aware framework that is well-suited for application in regions where tourism data is limited, such as Iraq, ultimately advancing tourism recommender systems by addressing their limitations in complex and data-scarce environments. © 2025, CC BY.

关键词： Tourism

来源：评论

学校读者我要写书评

暂无评论

Automatic Approach for Building Dataset of Citation Functions for COVID-19 Academic Papers 16

Automatic Approach for Building Dataset of Citation Function...

引用

16th Linguistic Annotation Workshop, LAW 2022

作者： Basuki, Setio Tsuchiya, Masatoshi Department of Computer Science and Engineering Toyohashi University of Technology 1-1 Hibarigaoka Tempaku-cho Aichi Toyohashi441-8580 Japan

ISBN: (纸本)9782493814081

This paper develops a new dataset of citation functions of COVID-19-related academic papers. Because the preparation of new labels of citation functions and building a new dataset requires much human effort and is time-consuming, this paper uses our previous citation functions that were built for the computer science (CS) domain, which consists of five coarse-grained labels and 21 fine-grained labels. This paper uses the COVID-19 Open Research Dataset (CORD-19) and extracts 99.6k random citing sentences from 10.1k papers. These citing sentences are categorized using the classification models built from the CS domain. The manually check on 475 random samples resulted accuracies of 76.6% and 70.2% on coarse-grained labels and fine-grained labels, respectively. The evaluation reveals three findings. First, two fine-grained labels experienced meaning shift while retaining the same idea. Second, the COVID-19 domain is dominated by statements highlighting the importance, cruciality, usefulness, benefit, consideration, etc. of certain topics for making sensible argumentation. Third, discussing State of The Arts (SOTA) in terms of their outperforming previous works in the COVID-19 domain is less popular compared to the CS domain. Our results will be used for further dataset development by classifying citing sentences in all papers from CORD-19. © 2022 European Language Resources Association (ELRA).

关键词： COVID-19

来源：评论

学校读者我要写书评

暂无评论

A Cross-Chain Interoperability Architecture for Smart City Environments

A Cross-Chain Interoperability Architecture for Smart City E...

引用

2022 IEEE GLOBECOM Workshops, GC Wkshps 2022

作者： Amet, Matthieu Darshan, M. Srivastava, Gautam Crichigno, Jorge Lakehead University Department of Computer Science Thunder BayONP7B 5E1 Canada Université de Lorraine Department of Computer Science Nancy France Amrita School of Engineering Department of Computer Science and Engineering Coimbatore India Brandon University Department of Mathematics and Computer Science BrandonMBR7A 6A9 Canada University of South Carolina College of Engineering and Computing Columbia United States

ISBN: (纸本)9781665459754

Blockchains have become quintessential for cutting-edge technology demands in the world. This trend will only continue to increase in the future. Looking at all modern technologies that seem to move towards decentralized architecture, blockchain and distributed ledger technology fit perfectly. In times of conventional database-orientated systems, modern frameworks made huge developments when application programming interfaces (APIs) and data were used cross-platform between centralized entities. A natural evolution for blockchain technology would be to enable communication and data exchange between blockchains (ie. both private and public). This could help not only digitize but could also set in a revolution for digital record-keeping that can be used for automation in the future. Our research work enables and tests interoperability between blockchains in different real-world scenarios. The research work also tries to understand various elements of a smart city environment and a few use-cases are discussed and experimented on. © 2022 IEEE.

关键词： Interoperability

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：