检索结果-内蒙古大学图书馆

Alignment efficient image-sentence retrieval considering transferable cross-modal representation learning

Frontiers of computer science 2024年第1期18卷 77-91页

作者： Yang YANG Jinyi GUO Guangyu LI Lanyu LI Wenjie LI Jian YANG School of Computer Science and Engineering Nanjing University of Science and TechnologyNanjing 210094China Department of Computing Hong Kong Polytechnic UniversityHong Kong 100872China State Key Lab.for Novel Software Technology Nanjing UniversityNanjing 210094China 14th Research Institute of China Electronics Technology Group Corporation Nanjing 210094China guangyu.li2017@***

Traditional image-sentence cross-modal retrieval methods usually aim to learn consistent representations of heterogeneous modalities,thereby to search similar instances in one modality according to the query from another modality in *** basic assumption behind these methods is that parallel multi-modal data(i.e.,different modalities of the same example are aligned)can be obtained in *** other words,the image-sentence cross-modal retrieval task is a supervised task with the alignments as ***,in many real-world applications,it is difficult to realign a large amount of parallel data for new scenarios due to the substantial labor costs,leading the non-parallel multi-modal data and existing methods cannot be used *** the other hand,there actually exists auxiliary parallel multi-modal data with similar semantics,which can assist the non-parallel data to learn the consistent ***,in this paper,we aim at“Alignment Efficient Image-Sentence Retrieval”(AEIR),which recurs to the auxiliary parallel image-sentence data as the source domain data,and takes the non-parallel data as the target domain *** single-modal transfer learning,AEIR learns consistent image-sentence cross-modal representations of target domain by transferring the alignments of existing parallel ***,AEIR learns the image-sentence consistent representations in source domain with parallel data,while transferring the alignment knowledge across domains by jointly optimizing a novel designed cross-domain cross-modal metric learning based constraint with intra-modal domain adversarial ***,we can effectively learn the consistent representations for target domain considering both the structure and semantic ***,extensive experiments on different transfer scenarios validate that AEIR can achieve better retrieval results comparing with the baselines.

关键词： image-sentence retrieval transfer learning semantic transfer structure transfer

来源：评论

学校读者我要写书评

暂无评论

Transformers in Small Object Detection: A Benchmark and Survey of State-of-the-Art

arXiv

引用

arXiv 2023年

作者： Rekavandi, Aref Miri Rashidi, Shima Boussaid, Farid Hoefs, Stephen Akbas, Emre Bennamoun, Mohammed Department of Computer Science and Software Engineering The University of Western Australia Australia Department of Electrical Electronics and Computer Engineering The University of Western Australia Australia Defence Science and Technology Group Australia Department of Computer Engineering Middle East Technical University Turkey

Transformers have rapidly gained popularity in computer vision, especially in the field of object recognition and detection. Upon examining the outcomes of state-of-the-art object detection methods, we noticed that transformers consistently outperformed well-established CNN-based detectors in almost every video or image dataset. While transformer-based approaches remain at the forefront of small object detection (SOD) techniques, this paper aims to explore the performance benefits offered by such extensive networks and identify potential reasons for their SOD superiority. Small objects have been identified as one of the most challenging object types in detection frameworks due to their low visibility. We aim to investigate potential strategies that could enhance transformers’ performance in SOD. This survey presents a taxonomy of over 60 research studies on developed transformers for the task of SOD, spanning the years 2020 to 2023. These studies encompass a variety of detection applications, including small object detection in generic images, aerial images, medical images, active millimeter images, underwater images, and videos. We also compile and present a list of 12 large-scale datasets suitable for SOD that were overlooked in previous studies and compare the performance of the reviewed studies using popular metrics such as mean Average Precision (mAP), Frames Per Second (FPS), number of parameters, and more. Researchers can keep track of newer studies on our web page, which is available at: https://***/arekavandi/Transformer-SOD. © 2023, CC BY-NC-SA.

关键词： Object recognition

来源：评论

学校读者我要写书评

暂无评论

A machine learning approach for skin lesion classification on iOS: implementing and optimizing a convolutional transfer learning model with Create ML

引用

International Journal of computers and Applications 2024年第8期46卷 666-685页

作者： Benedek Szabo, Aron Katona, Jozsef Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Department of Instrumentation and Automation Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary Department of Informatics GAMF Faculty of Engineering and Computer Science Kecskemét Hungary

The integration of machine learning (ML) into mobile applications presents unique challenges, particularly in resource-constrained environments such as iOS devices. Skin lesion classification is a critical task in dermatology, where accurate and efficient diagnostic tools can significantly aid in early detection of malignant lesions. This study aims to implement a machine learning-based iOS mobile application and develop a binary classification model for skin lesion images to determine whether a lesion is malignant. The research utilized Create ML to develop a convolutional neural network optimized for iOS, employing a transfer learning approach. A logistic regression model was cascaded with the convolutional neural network N to enhance classification accuracy. The model's performance was assessed through various validation metrics, ensuring its robustness and efficiency within the constraints of mobile hardware. A curated dataset from the International Skin Imaging Collaboration archive was used for training and testing. The model achieved an accuracy of 92%, a precision of 90%, a recall of 93%, and an F1-score of 91% in classifying skin lesions. These metrics validate the model's efficacy in identifying malignant lesions. Data curation involved collecting, labeling, and preparing a dataset from publicly available sources, ensuring the inclusion of diagnostically relevant features. The final model was integrated into an iOS application using Core ML and Vision frameworks. The developed application demonstrates reliable performance in classifying skin lesions with high accuracy on iOS devices. The inclusion of comprehensive performance metrics justifies the efficacy of the proposed approach. Future work will explore enhancements in model architecture and object detection capabilities to further improve diagnostic precision and application usability. © 2024 Informa UK Limited, trading as Taylor & Francis group.

关键词： Image recognition

来源：评论

学校读者我要写书评

暂无评论

Using analytic hierarchy process to evaluate deep learning for infrared target recognition

引用

Multimedia Tools and Applications 2024年第38期83卷 86229-86245页

作者： Xu, Laixiang Cao, Bingxu Wei, Xinhong Li, Bei Su, Jingfeng Ning, Shiyuan Zhao, Junmin School of Computer and Data Science Henan University of Urban Construction Pingdingshan467036 China School of Information Engineering Luohe Vocational Technology College Henan University of Technology Luohe Institute of Technology Luohe462000 China School of Computer and Data Science Innovation Laboratory of Smart Traffic and Big Data Development of Hainan Province Pingdingshan467036 China School of Computer and Data Science Research Center of Smart City and Big Data Engineering of Henan Province Henan University of Urban Construction Pingdingshan467036 China Department of Software Information China Electronics Technology Group Corporation 36-th Research Institute Jiaxing314033 China School of Computer and Data Science Research Center of Smart City and Big Data Engineering of Henan Province Research Center of Intelligent Campus Application Engineering of Henan Province Innovation Laboratory of Smart Traffic and Big Data Development of Hainan Province Henan University of Urban Construction Pingdingshan467036 China

In order to promote the evaluation performance of deep learning infrared automatic target recognition (ATR) algorithms in the complex environment of air-to-air missile research, we proposed an analytic hierarchy process (AHP) evaluation system consisting of the target recognition ability, the generalization recognition ability, the anti-interference ability, the background suppression ability, and the cost metrics ability. According to detailed quantitative analysis and experimental results, we establish probability evaluation models and get their probability values. Compared with traditional evaluation systems, our proposed method implementation is showing high reliability. Similar superior performances are also achieved, such as quick, comprehensive, and scientific. Through the above research, the key technology of deep learning performance evaluation in automatic target recognition has been broken through, which provides some basis and guidance for the analysis of infrared imaging-guided weapon systems. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Infrared imaging

来源：评论

学校读者我要写书评

暂无评论

A 32-Channel Time-to-Digital Converter with 20-ps Resolution for ToF Applications 4

A 32-Channel Time-to-Digital Converter with 20-ps Resolution...

引用

4th IEEE International Conference on Circuits and systems, ICCS 2022

作者： Ke, Qin Yu, Hang Li, Yan Ma, Siguang Xi, Daoming Chen, Rui Shenzhen University Shenzhen Laboratory of Ic Design for Iot College of Computer Science and Software Engineering Shenzhen China Physical Electronics Group of Jihua Laboratory Foshan China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Hefei China Huazhong University of Science and Technology Department of Biomedical Engineering Wuhan China

ISBN: (数字)9781665460361

ISBN: (纸本)9781665460361

Single-photon avalanche diodes (SPAD) enable sensitive detection of single photons under ultra-weak light conditions. Combined with direct time-of-flight (ToF) measurements, SPAD sensors are widely used in 3D imaging systems. Time-to-Digital Converter (TDC) can accurately measure time intervals in picoseconds, which is widely used for direct TOF measurements. This paper proposes a 32-channel TDC of two-level architecture containing a delay locked loop (DLL) and a vernier delay line (VDL) loop to increase timing accuracy. Designed in a 180 nm CMOS process, the TDC occupies an area of 4.1 mm × 3.6 mm and achieves 20 ps resolution at 2 Msps measurement rate. © 2022 IEEE.

关键词： Particle beams

来源：评论

学校读者我要写书评

暂无评论

AI-based Adaptive Programming Education for Socially Disadvantaged Students: Bridging the Digital Divide

引用

TechTrends 2025年 1-18页

作者： Katona, Jozsef Gyonyoru, Klara Ida Katonane Department of Software Development and Application Institute of Computer Engineering University of Dunaujvaros Dunaujvaros Hungary Institute of Electronics and Communication Systems Kandó Kálmán Faculty of Electrical Engineering Obuda University Budapest Hungary GAMF Faculty of Engineering and Computer Science John Von Neumann University Kecskemet Hungary Department of Applied Quantitative Methods Faculty of Finance and Accountancy Budapest University of Economics and Business Budapest Hungary Department of Organizational Development and Communication Science Institute of Social Sciences University of Dunaujvaros Dunaujvaros Hungary

In the context of the digital economy, programming proficiency is an essential competency that promotes upward socio-economic mobility and expands career opportunities. However, students from socially disadvantaged backgrounds often face significant barriers to acquiring these skills, such as limited access to technology and educational resources. This study explores the impact of AI-based adaptive programming education on socially disadvantaged students' learning outcomes and engagement levels. The 122 participants in the research were divided into an experimental group (EG) that received AI-driven adaptive instruction, and a control group (CG) taught through the traditional curriculum during the 13-week experimental period. In the research, the combined pre-test/post-test assessments and self-report engagement questionnaires were used to focus on behavioural, emotional, and cognitive engagement. The findings showed that the EG demonstrated higher levels in programming knowledge and full-scale engagement across behavioural, emotional, and cognitive dimensions compared to the CG. This difference was confirmed through ANOVA, while ANCOVA, which controlled for students' socio-economic factors, further confirmed that the AI system had a positive impact on students’ results regardless of their socio-economic background. Indeed, current research findings suggest that AI-based adaptive learning environments can hold a promising opportunity to narrow educational gaps, boost engagement, and contribute toward better learning outcomes for socially disadvantaged students. © The Author(s) 2025.

关键词： Adaptive learning AI-based education Disadvantaged students Programming education Student engagement

来源：评论

学校读者我要写书评

暂无评论

Game-Oriented Security Strategy Against Hotspot Attacks for Internet of Vehicles

引用

computers, Materials & Continua 2021年第8期68卷 2145-2157页

作者： Juan Guo Yanzhu Liu Shan Li Zhi Li Sonia Kherbachi Guilin University of Electronic Technology Beihai536000GuangxiChina Great Wall Computer Software and Systems Inc. Beijing100083China Department of Electronics and Communication Engineering Beijing Electronic Science and Technology InstituteBeijing100070China State Grid Information&Telecommunication Group Co. Ltd.Beijing100031China Department of Management University of BejaiaBejaia06000Algeria

With the rapid development of mobile communication technology,the application of internet of vehicles(IoV)services,such as for information services,driving safety,and traffic efficiency,is growing *** businesses with low transmission delay,high data processing capacity and large storage capacity,by deploying edge computing in the IoV,data processing,encryption and decision-making can be completed at the local end,thus providing real-time and highly reliable communication *** roadside unit(RSU),as an important part of edge computing in the IoV,fulfils an important data forwarding function and provides an interactive communication channel for vehicles and server *** computing resources can be configured to accommodate the computing requirements of *** this study,a virtual traffic defense strategy based on a differential game is proposed to solve the security problem of user-sensitive information leakage when an RSU is *** incentive mechanism encourages service vehicles within the hot range to send virtual traffic to another *** attracting the attention of attackers,it covers the target RSU and protects the system from *** results show that the scheme provides the optimal strategy for intelligent vehicles to transmit virtual data,and ensures the maximization of users’interests.

关键词： Edge computing internet of vehicles differential games security defense

来源：评论

学校读者我要写书评

暂无评论

Integrating AI-based adaptive learning into the flipped classroom model to enhance engagement and learning outcomes

引用

computers and Education: Artificial Intelligence 2025年 8卷

This study explores the integration of the AI-powered adaptive learning system within the flipped classroom model and discusses programming education. The increasing demand for the acquisition of programming skills in the digital era generates the development of innovative pedagogical approaches to help students overcome the difficulties in mastering programming concepts. The study investigates how AI-based adaptive feedback influences student engagement, motivation, and learning outcomes. This study was conducted for 13 weeks, with two groups of undergraduate students. In one of the groups, tshe traditional flipped classroom model was applied, while in the other group, AI-driven adaptive learning tools were introduced. In this research methodology, a mixed-methods approach was adopted, integrating quantitative analyses with qualitative data. These showed that the experimental group demonstrated significant improvements in learning results and motivation, thereby realiszing the value of real-time, personalized feedback provided through the AI system. The qualitative findings revealed that with the AI system, there was an increase in student autonomy;thus, there is increased responsibility among students to collaborate in in-class participation, which helps enhance a more dynamic and cooperative learning environment. On the whole, these findings reveal important implications of the integration of AI into flipped classrooms, while at the same time, they reveal how to make the modernization of programming education achieve a better overall experience in student learning. © 2025 The Authors

关键词： Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

The effects of Tx./Rx. pointing errors on the performance efficiency of local area optical wireless communication networks

引用

Journal of Optical Communications 2024年第s1期45卷 s253-s259页

作者： Eid, Mahmoud M. A. Zaki Rashed, Ahmed Nabih Ahammad, Mohd. Sultan Paul, Bikash Kumar Ahmed, Kawsar Department of Electrical Engineering College of Engineering Taif University Taif Saudi Arabia Electronics and Electrical Communications Engineering Department Faculty of Electronic Engineering Menoufia University Menouf32951 Egypt Department of Computer Science and Engineering Mawlana Bhashani Science and Technology University Santosh Tangail1902 Bangladesh Department of Information and Communication Technology Mawlana Bhashani Science and Technology University Santosh Tangail1902 Bangladesh Group of Bio-photomatiχ Mawlana Bhashani Science and Technology University Santosh Tangail1902 Bangladesh Department of Software Engineering Daffodil International University Dhanmondi Sukrabad Dhaka1207 Bangladesh

This study presents the effects of Tx./Rx. pointing errors on the performance efficiency of local area optical wireless communication networks. The received signal power and max Q factor are measured in the presence of vertical-cavity surface-emitting lasers (VCSELs) bias current and modulation current at maximum propagation distance for the wireless network. The signal is enhanced with increasing of bias and modulation peak currents of the VCSEL device. The optimum received power and max. Q factor is also examined at Tx./Rx. the pointing error of 0.1 mrad and propagation reach of 5 km at available bit rates transmission of 10 Gb/s. © 2021 Walter de Gruyter GmbH, Berlin/Boston.

关键词： Surface emitting lasers

来源：评论

学校读者我要写书评

暂无评论

Conventional/Phase Shift Dual Drive Mach-Zehnder Modulation Measured Type Based Radio over Fiber systems

引用

Journal of Optical Communications 2024年第s1期45卷 s123-s131页

作者： Amiri, I.S. Rashed, Ahmed Nabih Zaki Rahman, Ziaur Paul, Bikash Kumar Ahmed, Kawsar Computational Optics Research Group Advanced Institute of Materials Science Ton Duc Thang University Ho Chi Minh City Viet Nam Faculty of Applied Sciences Ton Duc Thang University Ho Chi Minh City Viet Nam Electronics and Electrical Communications Engineering Department Faculty of Electronic Engineering Menoufia University Menouf32951 Egypt School of Computer Science and Software Engineering RMIT University 414-418 Swanston Street MelbourneVIC3000 Australia Department of Information and Communication Technology Mawlana Bhashani Science and Technology University Santosh Tangail1902 Bangladesh Group of Bio-photomatiχ Mawlana Bhashani Science and Technology University Santosh Tangail1902 Bangladesh Department of Software Engineering Daffodil International University Sukrabad Dhanmondi Dhaka1207 Bangladesh

This work clarified the conventional/phase shift dual Mach-Zehnder modulation measured technique for upgrading radio over fiber (ROF) systems. Four levels of pulse amplitude modulation and non-return to zero code scheme are employed in different previous studies for upgrading ROF systems. The possible transmission distance is extended to 150 km and a bit rate of 40 Gb/s for achieved acceptable max. Q factor of 13.7498 and minimum bit error rates of 1.82×10-43. The total optical received power is measured at fiber cable length for maximum distance. Also, in the same way, the signal power levels are measured after fiber optic cable and avalanche photo detector optical receiver. The proposed model has presented the enhancement percentage ratio of 13.654% over the previous models in signal quality enhancement. © 2020 Walter de Gruyter GmbH, Berlin/Boston.

关键词： Q factor measurement

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：