检索结果-内蒙古大学图书馆

10th International Conference on engineering and Emerging Technologies, ICEET 2024

作者： Hassan, Aqib Memon, Vengas Shaikh, Farhan Bashir Alahmari, Saad Iqbal, Mansoor Azam, Farooq School of Computing and Mathematical Sciences University of Greenwich London United Kingdom Universiti Tunku Faculty of Information and Communication Tech. Abdul Rahman Perak Malaysia Northern Border University Department of Computer Science Arar Saudi Arabia University of Wah Department of Computer Science Wah Cantt Pakistan Nust Dept. of Computer Software Engineering Islamabad Pakistan

ISBN: (纸本)9798331532895

The rise in demand for real-time applications, such as live streaming, online gaming, and Internet telephony, has highlighted the necessity for transport protocols that offer low latency and network stability. Traditional protocols, namely Transmission Control Protocol (TCP) and User Datagram Protocol (UDP), have inherent limitations. TCP ensures reliable data delivery but suffers from latency, making it unsuitable for real-time needs, while UDP provides fast data transmission but lacks congestion control, which can lead to network instability. This gap underscores the need for a protocol that combines high-speed data transfer with effective congestion management. To address this issue, the Internet engineering Task Force (IETF) introduced the Datagram Congestion Control Protocol (DCCP), designed to optimize data delivery speed while managing congestion. This study evaluates DCCP's performance against TCP in network scenarios of 8 and 16 nodes using the NS2 simulation environment. By comparing throughput and delay across various bandwidth and latency settings, the study finds that DCCP offers up to 35% higher throughput in 8-node setups and reduces packet loss by 20% in high-bandwidth scenarios. These results suggest that DCCP is well-suited for real-time applications, achieving a crucial balance between fast data transmission and network stability. © 2024 IEEE.

关键词： Congestion control (communication)

来源：评论

学校读者我要写书评

暂无评论

An Enhanced Authentication Protocol Suitable for Constrained RFID Systems

引用

IEEE Access 2024年 12卷 61610-61628页

作者： Hosseinzadeh, Mehdi Servati, Mohammad Reza Rahmani, Amir Masoud Safkhani, Masoumeh Lansky, Jan Janoscova, Renata Ahmed, Omed Hassan Tanveer, Jawad Lee, Sang-Woong Duy Tan University Institute of Research and Development Da Nang550000 Viet Nam Duy Tan University School of Medicine and Pharmacy Da Nang550000 Viet Nam Shahid Rajaee Teacher Training University Department of Computer Engineering Tehran16788-15811 Iran National Yunlin University of Science and Technology Future Technology Research Center Douliou Yunlin64002 Taiwan School of Computer Science Tehran19395-5746 Iran University of Finance and Administration Faculty of Economic Studies Department of Computer Science and Mathematics Prague101 00 Czech Republic University of Human Development Department of Information Technology Sulaymaniyah0778-6 Iraq Sejong University Department of Computer Science and Engineering Seoul05006 Korea Republic of Gachon University Pattern Recognition and Machine Learning Laboratory Department of Ai Software Seongnam-si13557 Korea Republic of

RFID technology offers an affordable and user-friendly solution for contactless identification of objects and individuals. However, the widespread adoption of RFID systems raises concerns regarding security and privacy. Vulnerabilities such as message tampering, interception, and eavesdropping pose significant risks to the integrity of the system. This study examines the effectiveness of two recently proposed ultra-lightweight RFID authentication protocols, URASP and KUAJB, in addressing these security challenges. Surprisingly, our findings reveal that both URASP and KUAJB protocols are susceptible to secret disclosure attacks, despite their claims of providing robust security. Consequently, the development of a truly ultra-lightweight protocol that ensures adequate security becomes a formidable task. As a potential solution, the enhancement of the URASP protocol through the integration of a simple cryptographic primitive is suggested to bolster its security measures. © 2013 IEEE.

关键词： Hash functions

来源：评论

学校读者我要写书评

暂无评论

MedCLIP-SAM: Bridging Text and Image Towards Universal Medical Image Segmentation

arXiv

引用

arXiv 2024年

作者： Koleilat, Taha Asgariandehkordi, Hojat Rivaz, Hassan Xiao, Yiming Department of Electrical and Computer Engineering Concordia University Montreal Canada Department of Computer Science and Software Engineering Concordia University Montreal Canada

Medical image segmentation of anatomical structures and pathology is crucial in modern clinical diagnosis, disease study, and treatment planning. To date, great progress has been made in deep learning-based segmentation techniques, but most methods still lack data efficiency, generalizability, and interactability. Consequently, the development of new, precise segmentation methods that demand fewer labeled datasets is of utmost importance in medical image analysis. Recently, the emergence of foundation models, such as CLIP and Segment-Anything-Model (SAM), with comprehensive cross-domain representation opened the door for interactive and universal image segmentation. However, exploration of these models for data-efficient medical image segmentation is still limited, but is highly necessary. In this paper, we propose a novel framework, called MedCLIP-SAM that combines CLIP and SAM models to generate segmentation of clinical scans using text prompts in both zero-shot and weakly supervised settings. To achieve this, we employed a new Decoupled Hard Negative Noise Contrastive Estimation (DHN-NCE) loss to fine-tune the BiomedCLIP model and the recent gScoreCAM to generate prompts to obtain segmentation masks from SAM in a zero-shot setting. Additionally, we explored the use of zero-shot segmentation labels in a weakly supervised paradigm to improve the segmentation quality further. By extensively testing three diverse segmentation tasks and medical image modalities (breast tumor ultrasound, brain tumor MRI, and lung X-ray), our proposed framework has demonstrated excellent accuracy. Code is available at https://***/HealthX-Lab/MedCLIP-SAM. © 2024, CC BY-NC-SA.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Recognition and Detection of Diabetic Retinopathy Using Densenet-65 Based Faster-RCNN

引用

computers, Materials & Continua 2021年第5期67卷 1333-1351页

作者： Saleh Albahli Tahira Nazir Aun Irtaza Ali Javed Department of Information Technology College of ComputerQassim UniversityBuraydahSaudi Arabia Department of Computer Science University of Engineering and TechnologyTaxila47050Pakistan Department of Software Engineering University of Engineering and TechnologyTaxila47050Pakistan

Diabetes is a metabolic disorder that results in a retinal complication called diabetic retinopathy(DR)which is one of the four main reasons for sightlessness all over the *** usually has no clear symptoms before the onset,thus making disease identication a challenging *** healthcare industry may face unfavorable consequences if the gap in identifying DR is not lled with effective ***,our objective is to develop an automatic and cost-effective method for classifying DR *** this work,we present a custom Faster-RCNN technique for the recognition and classication of DR lesions from retinal *** pre-processing,we generate the annotations of the dataset which is required for model ***,introduce DenseNet-65 at the feature extraction level of Faster-RCNN to compute the representative set of key ***,the Faster-RCNN localizes and classies the input sample into ve *** experiments performed on a Kaggle dataset comprising of 88,704 images show that the introduced methodology outperforms with an accuracy of 97.2%.We have compared our technique with state-of-the-art approaches to show its robustness in term of DR localization and ***,we performed cross-dataset validation on the Kaggle and APTOS datasets and achieved remarkable results on both training and testing phases.

关键词： Deep learning medical informatics diabetic retinopathy healthcare computer vision

来源：评论

学校读者我要写书评

暂无评论

AI-Driven Innovations: Transforming Data Support and Environmental Management for the Future

AI-Driven Innovations: Transforming Data Support and Environ...

引用

Decision Aid sciences and Application (DASA), International Conference on

作者： Mahmoud Mahfuri Amer Ibrahim Computer and information Technology Department International University for Science and Technology Department of Computer Science and Software Engineering United Arab Emirates University

ISBN: (数字)9798350369106

ISBN: (纸本)9798350369113

This paper presents the evolving role of artificial intelligence (AI) in improving internal control and management processes. AI-driven technologies, including Generative Adversarial Networks (GANs) and ontologies, in addition to enabling data-supported decision-making, and empowering local governments to execute administrative tasks more efficiently and effectively. The study explores the wide range of opportunities that AI offers, highlighting the potential for optimizing internal workflows. Besides, it delves into key challenges, models, and methods for fully integrating AI into organizational processes, providing actionable insights for achieving streamlined, data-driven management systems.

关键词： Technological innovation Local government Decision making Process control Ontologies Generative adversarial networks Artificial intelligence Environmental management

来源：评论

学校读者我要写书评

暂无评论

Efficient Zero-Shot Voice Cloning for Bengali Speech Synthesis

Efficient Zero-Shot Voice Cloning for Bengali Speech Synthes...

引用

International Conference on computer and Information Technology (ICCIT)

作者： Abir Ahmed Prapti Roy Durjoy Chandra Paul M. Shahidur Rahman Department of Software Engineering Shahjalal University of Science & Technology Sylhet Bangladesh Department of Computer Science & Engineering Shahjalal University of Science & Technology Sylhet Bangladesh Department of Computer Science & Engineering Sylhet Engineering College Sylhet Bangladesh

ISBN: (数字)9798331519094

ISBN: (纸本)9798331519100

Voice cloning has numerous useful applications, including assisting individuals who have lost their ability to speak, movie dubbing, and translating voices into different languages. However, voice cloning in Bangla is still in its early stages. In this work, we explore two standard approaches for zero-shot Bengali voice cloning, which can generate cloned audio from short samples, even for Bengali speakers not included in the training data. Despite using limited datasets, our models achieve cloning quality comparable to state-of-the-art models trained on much larger datasets, maximizing potential benefits for the Bengali community. We use a neural network technique called the speaker encoder (SE) voice cloning method, which achieved a Mean Opinion Score (MOS) of 3.8 and 83% cosine similarity between the original and cloned voice. This model was trained on a custom multi-speaker dataset. Additionally, we employ a speaker converter (SC) approach trained on a single-speaker dataset, which achieved a MOS of 4.0 and 82.64% cosine similarity. By combining the two models in an ensemble, we further improved performance, reaching an MOS of 4.4 and a cosine similarity of 85.21%. Our proposed approach can be an important step toward developing voice cloning technology for the Bengali language, potentially positively impacting the community. The demos can be accessed here.

关键词： Training Translation Neural networks Cloning Training data Motion pictures Ensemble learning Speech synthesis Information technology Standards

来源：评论

学校读者我要写书评

暂无评论

Exploring the Behavior Patterns of Students Accessing Online Learning Material in Online Course: A Case Study at Hung Vuong University 30

Exploring the Behavior Patterns of Students Accessing Online...

引用

30th International Conference on computers in Education Conference, ICCE 2022

作者： Tho, Pham-Duc Lai, Chih-Hung Department of Software Engineering Hung Vuong University Viet Nam Department of Computer Science and Information Engineering National Dong Hwa University Taiwan

ISBN: (纸本)9786269689002

Understanding students' behavior in online courses may provide teachers with useful information to improve their educational design and provide insights for content and instructional designers to develop personalized learning support. This research uses cluster analysis to explore learners' interaction with online learning materials behavior in an online course at Hung Vuong University, Vietnam and identified three clusters (Less-engaged students, Moderately-engaged students, and Highly-engaged students) which evince different behavior patterns with regards to the time spent interacting with various resources. Based on the findings, several suggestions are also proposed for future research. © ICCE *** rights reserved.

关键词： Cluster analysis

来源：评论

学校读者我要写书评

暂无评论

Secrecy Maximization for Pico Edge Users in 5G Backhaul HWNs: A Quantum RL Approach

Secrecy Maximization for Pico Edge Users in 5G Backhaul HWNs...

引用

2022 IEEE International Conference on Advanced Networks and Telecommunications Systems, ANTS 2022

作者： Sharma, Himanshu Pulkit Sharma, Gitika Kumar, Neeraj Thapar Institute of Engineering and Technology Computer Science and Engineering Department Punjab Patiala India MAQ Software Hyderabad India Bennett University School of Computer Science Engineering and Technology Greater Noida India

ISBN: (纸本)9781665473408

Picocells can enhance the capacity and spectral efficiency of heterogeneous wireless networks (HWNs) and reduce user equipment (UE) power consumption. Also, the wireless backhaul technology has emerged as an efficient way to boost the data rate in dense 5G HWNs. However, due to the dynamic and distributed nature of 5G HWNs, pico edge users (PEUs) cannot resist the eavesdropping attempts. Moreover, the PEUs are located far away from the PBS and receive considerably low signal-to-interference-plus-noise ratio (SINR) values. Thus, it is critical to improve the achievable secrecy rate (SR) of PEUs for reliable and secure transmission of confidential data in the existence of eavesdroppers. Aiming to enhance the SR of PEUs, this paper formulates a joint optimization problem of beamforming and power allocation of millimeter-wave (mmWave)enabled pico base station (PBS) having backhaul wireless links with macro base station (MBS). As the system is diverse and complex, addressing this non-convex optimization problem directly is challenging. Thus, we have used the Markov decision process (MDP) to transform this optimization problem into a multi-agent reinforcement learning (MARL) problem. Then, to determine the optimal policy for MARL problem of beamforming and power allocation, we propose a Quantum reinforcement learning (QRL) approach. It uses quantum parallelism and state superposition principles to maximize the SR of PEUs. Simulation results indicate that the proposed scheme achieves 28.12% and 7.9% better secrecy rate than MAQL and MADQL schemes, respectively. © 2022 IEEE.

关键词： Markov processes

来源：评论

学校读者我要写书评

暂无评论

European Perceptions of Artificial Intelligence and Their Social Variability. An Exploratory Study 24

European Perceptions of Artificial Intelligence and Their So...

引用

24th International Conference on Control Systems and computer science, CSCS 2023

作者： Budeanu, Andrea-Mariana Turcanu, Dinu Rosner, Daniel Tech Lounge Association Bucharest Romania Technical University of Moldova Department of Software Engineering and Automatics Chisinǎu Moldova University Politehnica of Bucharest Computer Science Department Bucharest Romania

ISBN: (纸本)9798350313390

We are currently living in societies that are profoundly concerned about the impact of current and potential technologies on our present and future lives. How is the future of Artificial Intelligence (AI) impact perceived in the general population of the European Union? How are its effects evaluated as regards the job markets? These are just two questions that require, from the general public, an exercise of information and also an exercise of the imagination. A survey is a useful instrument to collect these public perceptions for in-depth analysis of the respective societies. This paper is based on a secondary study of a 2021 Eurobarometer survey that provides insights into the perceived future impact of AI and the social sources of its variability. It reveals evolving aspects about AI's perception that help us better understand ourselves. The identified trends, relationships, as well as the lack of some expected correlations, offer useful information for various stakeholders interested in the social life of AI. © 2023 IEEE.

关键词： Artificial intelligence

来源：评论

学校读者我要写书评

暂无评论

Joint enhancement of automatic chest X-ray diagnosis and radiological gaze prediction with multi-stage cooperative learning

arXiv

引用

arXiv 2024年

作者： Qiu, Zirui Rivaz, Hassan Xiao, Yiming Department of Computer Science and Software Engineering Concordia University Montreal Canada Department of Electrical and Computer Engineering Concordia University Montreal Canada

Background: As visual inspection is an inherent process during radiological screening, the associated eye gaze data can provide valuable insights into relevant clinical decision processes and facilitate computer-assisted diagnosis. However, the relevant techniques are still under-explored. Purpose: With deep learning becoming the state-of-the-art for computer-assisted diagnosis, integrating human behavior, such as eye gaze data, into these systems is instrumental to help guide machine predictions with clinical diagnostic criteria, thus enhancing the quality of automatic radiological diagnosis. In addition, the ability to predict a radiologist’s gaze saliency from a clinical scan along with the automatic diagnostic result could be instrumental for the end users. Methods: We propose a novel deep learning framework for joint disease diagnosis and prediction of corresponding radiological gaze saliency maps for chest X-ray scans. Specifically, we introduce a new dual-encoder multi-task UNet, which leverages both a DenseNet201 backbone and a Residual and Squeeze-and-Excitation block-based encoder to extract diverse features for visual saliency map prediction, and a multi-scale feature-fusion classifier to perform disease classification. To tackle the issue of asynchronous training schedules of individual tasks in multi-task learning, we propose a multi-stage cooperative learning strategy, with contrastive learning for feature encoder pretraining to boost performance. Results: Our proposed method is shown to significantly outperform existing techniques for chest radiography diagnosis (AUC = 0.93) and the quality of visual saliency map prediction (correlation coefficient = 0.58). Conclusion: Benefiting from the proposed multi-task multi-stage cooperative learning, our technique demonstrates the benefit of integrating clinicians’ eye gaze into radiological AI systems to boost performance and potentially explainability. © 2024, CC BY-NC-SA.

关键词： Contrastive Learning

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：