检索结果-内蒙古大学图书馆

Global Power, Energy and Communication Conference (GPECOM)

作者： Suganiya Murugan Harish Krishnadas S Abiishek G Pradeep Kumar S Rubini B Department of Computing Technologies SRM Institute of Science and Technology Chennai India Department of Computer Science and Engineering SRM Institute of Science and Technology Chennai India Department of Electrical and Electronics Engineering Vels Institute of Science Technology and Advanced Studies Chennai India

ISBN: (数字)9798350351088

ISBN: (纸本)9798350351095

the Next Gen Driver Monitoring System utilizes advanced techniques to enhance road safety by detecting driver drowsiness, nodding, and blink patterns. Integrating cutting-edge technology, the system analyzes facial landmarks and eye movements in real-time through sophisticated algorithms. Through the fusion of computer vision and advanced algorithms, it accurately identifies signs of driver fatigue and distraction, providing timely alerts to mitigate potential risks on the road. This research delves into the design, implementation, and evaluation of the system, demonstrating its effectiveness in reducing accidents caused by drowsy or inattentive driving. The Next Gen Driver Monitoring System represents a significant advancement in automotive safety technology, ushering in safer driving experiences.

关键词： computer vision Vehicle safety Fatigue Road safety Real-time systems Monitoring Vehicles

来源：评论

学校读者我要写书评

暂无评论

Diabetic Retinopathy Detection Using Statistical Features 2

Diabetic Retinopathy Detection Using Statistical Features

引用

2nd International Conference on Innovative Practices in Technology and Management, ICIPTM 2022

作者： Reethika, A. Sathish, J. Kanaga Priya, P. Shadrach, Finney Daniel Kanivarshini, M.S. Kpr Institute Of Engineering And Technology Department Of Electronics And Communication Engineering Tamil Nadu Coimbatore India Dr. N.G.P Institute Of Technology Department Of Electrical And Electronics Engineering Tamil Nadu Coimbatore India Kpr Institute Of Engineering And Technology Department Of Computer Science And Engineering Tamil Nadu Coimbatore India Hcl Technologies Tamil Nadu Chennai India

ISBN: (纸本)9781665466431

Diabetic Retinopathy is a medical condition that occurs due to prolonged diabetes, ensuing in harmful to small blood vessels present in the retina, which in later stage leads to complete loss of vision. Moreover, this disease is asymptomatic at its earlier stages and hence, constant screening and precise diagnosis of Diabetic Retinopathy is required. Fundus image analysis proves to be an indispensable tool for the diagnosis of patients. This Paper focuses on using a neural network algorithm aimed to diagnosis of the disease by extracting the key features in the retinal fundus images are lesions, hemorrhages, exudates then the statistical parameters which henceforth help in the classification. This algorithm proves to achieve better accuracy than the traditional unsupervised classification algorithms. © 2022 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Electronic Preresonance Stimulated Raman Scattering Spectromicroscopy Using Multiple-Plate Continuum

引用

Journal of Physical Chemistry B 2023年第31期127卷 6896-6902页

作者： Huang, Guan-Jie Li, Cheng-Wei Lee, Po-Yi Su, Jia-Xuan Chao, Kuo-Chuan Chu, Li-An Chiang, Ann-Shyn Cheng, Ji-Xin Chen, Bo-Han Lu, Chih-Hsuan Chu, Shi-Wei Yang, Shang-Da Department of Physics National Taiwan University Taipei10617 Taiwan Institute of Photonics Technologies National Tsing Hua University Hsinchu300044 Taiwan Brain Research Center National Tsing Hua University Hsinchu300044 Taiwan Department of Biomedical Engineering & Environmental Sciences National Tsing Hua University Hsinchu30044 Taiwan Institute of Systems Neuroscience and Department of Life Science National Tsing Hua University Hsinchu30044 Taiwan Department of Electrical and Computer Engineering Boston University BostonMA02215 United States Molecular Imaging Center National Taiwan University Taipei10617 Taiwan

Stimulated Raman scattering (SRS) spectromicroscopy is a powerful technique that enables label-free detection of chemical bonds with high specificity. However, the low Raman cross section due to typical far-electronic resonance excitation seriously restricts the sensitivity and undermines its application to bio-imaging. To address this bottleneck, the electronic preresonance (EPR) SRS technique has been developed to enhance the Raman signals by shifting the excitation frequency toward the molecular absorption. A fundamental weakness of the previous demonstration is the lack of dual-wavelength tunability, making EPR-SRS only applicable to a limited number of species in the proof-of-concept experiment. Here, we demonstrate the EPR-SRS spectromicroscopy using a multiple-plate continuum (MPC) light source able to examine a single vibration mode with independently adjustable pump and Stokes wavelengths. In our experiments, the C═C vibration mode of Alexa 635 is interrogated by continuously scanning the pump-to-absorption frequency detuning throughout the entire EPR region enabled by MPC. The results exhibit 150-fold SRS signal enhancement and good agreement with the Albrecht A-term preresonance model. Signal enhancement is also observed in EPR-SRS images of the whole Drosophila brain stained with Alexa 635. With the improved sensitivity and potential to implement hyperspectral measurement, we envision that MPC-EPR-SRS spectromicroscopy can bring the Raman techniques closer to a routine in bio-imaging. © 2023 American Chemical Society.

关键词： Stimulated Raman scattering

来源：评论

学校读者我要写书评

暂无评论

Lfg: An Easy-to-Use Realistic Synthetic Landfill Generator

SSRN

引用

SSRN 2024年

作者： Petsanis, Thanos Kapoutsis, Athanasios Ch. Kosmatopoulos, Elias B. Department of Electrical and Computer Engineering Democritus University of Thrace Xanthi67100 Greece Information Technologies Institute The Centre for Research & Technology Hellas Thessaloniki57001 Greece

In this paper we present a Blender add-on named LFG that allows for easy, large and realistic, 3D model LandFill Generation. Large datasets of vast, diverse synthetic landfills are hard to come by, and greatly in need for the purposes of developing and evaluating a multitude of algorithms (e.g. waste classification, 3D-reconstruction, volume estimation algorithms) in the context of research against environment crime. Additionally, they can be used alongside UAV simulators for the development of path-planning algorithms. Although there are some 3D models of landfills available on online 3D-model marketplaces, these are often expensive, low-quality, low-variety and unalterable models. LFG offers customizable, expandable options and realistic features tailored for landfill generation and research. © 2024, The Authors. All rights reserved.

关键词： 3D modeling

来源：评论

学校读者我要写书评

暂无评论

Test-Time Training-Free Domain Adaptation 48

Test-Time Training-Free Domain Adaptation

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Feng, Yongxiang He, Weihua You, Kaichao Liu, Bing Zhang, Ziyang Wang, Yaoyuan Li, Minglei Lou, Yihang Li, Jiawei Li, Guoqi Liao, Jianxing Huawei Technologies Co. Ltd. Advanced Computing and Storage Lab China Peking University School of Electronics Engineering and Computer Science China Huawei Technologies Co. Ltd. Language & Speech Innovation Lab China Huawei Technologies Co. Ltd. GoTen AI Lab Department of Intelligent Vision China Huawei Technologies Co. Ltd. Department of Production Automation Development China Chinese Academy of Sciences Institute of Automation China

ISBN: (纸本)9781728163277

Deploying deep learning models to new environments is very challenging. Domain adaptation (DA) is a promising paradigm to solve the problem by collecting and adapting to unlabeled data in new environments. Though research efforts have led to steady performance improvement over the past decade, DA algorithms are still hard to deploy, as training on unlabeled new data makes tuning difficult and not feasible for inference-only devices. To make DA practical, in this paper we study a new problem named Test-time Training-Free Domain Adaptation (TTDA), where trained models must adapt to a single input (mimicking the test-time scenario) without training. By exploiting spatial activation that was previously overlooked and simply averaged out, we propose a simple method based on Feature Statistics Transformation (FST) on-the-fly for each test example. The proposed algorithm is tested in the TTDA setting on two standard DA benchmarks. Surprisingly, it surpasses or performs on par with state-of-the-art DA methods, even though they require additional training. We envision that this training-free paradigm has the potential to bring DA to embedded devices and would be of interest to audience of community. © 2023 IEEE.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Calibration Attacks: A Comprehensive Study of Adversarial Attacks on Model Confidence

arXiv

引用

arXiv 2024年

作者： Obadinma, Stephen Zhu, Xiaodan Guo, Hongyu Department of Electrical and Computer Engineering Ingenuity Labs Research Institute Queen’s University Canada Digital Technologies Research Centre National Research Council Canada

In this work, we highlight and perform a comprehensive study on calibration attacks, a form of adversarial attacks that aim to trap victim models to be heavily miscalibrated without altering their predicted labels, hence endangering the trustworthiness of the models and follow-up decision making based on their confidence. We propose four typical forms of calibration attacks: underconfidence, overconfidence, maximum miscalibration, and random confidence attacks, conducted in both black-box and white-box setups. We demonstrate that the attacks are highly effective on both convolutional and attention-based models: with a small number of queries, they seriously skew confidence without changing the predictive performance. Given the potential danger, we further investigate the effectiveness of a wide range of adversarial defence and recalibration methods, including our proposed defences specifically designed for calibration attacks to mitigate the harm. From the ECE and KS scores, we observe that there are still significant limitations in handling calibration attacks. To the best of our knowledge, this is the first dedicated study that provides a comprehensive investigation on calibration-focused attacks. We hope this study helps attract more attention to these types of attacks and hence hamper their potential serious damages. To this end, this work also provides detailed analyses to understand the characteristics of the attacks. Copyright © 2024, The Authors. All rights reserved.

关键词： Generative adversarial networks

来源：评论

学校读者我要写书评

暂无评论

A Study of Discrete Multitone Modulation for Wireline Links beyond 100 Gb/s

IEEE Open Journal of Circuits and Systems

引用

IEEE Open Journal of Circuits and Systems 2021年 2卷 78-90页

作者： Vatankhahghadim, Behraz Wary, Nijwm Bailey, James Chan Carusone, Anthony Edward S. Rogers Sr. Department of Electrical and Computer Engineering University of Toronto Toronto Canada School of Electrical Science Indian Institute of Technology Bhubaneswar Bhubaneswar India Huawei Technologies Canada Markham Canada

To overcome the severe losses beyond 28 GHz in low-cost electrical channels, 4-level pulse-amplitude modulation (4-PAM) wireline links targeting 112 Gb/s incorporate resource- or power-intensive equalization schemes such as decision-feedback equalizers (DFE) with many taps. Alleviating the timing constraints that cause DFEs to balloon in size and power, discrete multitone (DMT) modulation involves independent sub-channels that can be equalized in the frequency domain without feedback. DMT allows flexibility in assigning bits to each sub-channel, thereby potentially avoiding lossy parts of the frequency spectrum. This article presents behavioural modeling results and an experimental setup used to study DMT transceivers. Our simulations show 200 Gb/s operation at a bit error rate (BER) of less than 10-5 over an IEEE P802.3ck channel with 21 dB of loss at 50 GHz and assuming 150-rms of jitter, 1.26mV_rms of noise at the receiver's input, and 7-bit 80 GS/s data converters. An updated bit-loading algorithm led to BER values 1-2 orders of magnitude below our previous results. The estimated power and area of the associated digital signal processing are comparable to those of DFE. We also present an experimental DMT setup achieving 61.6 Gb/s at a BER of 8.6 10-4 over a physical channel with severe notches that is very challenging for 4-PAM. © 2020 IEEE.

关键词： Decision feedback equalizers

来源：评论

学校读者我要写书评

暂无评论

The Technical Ethics used in Bad and SD Methods: A Deep Review 4

The Technical Ethics used in Bad and SD Methods: A Deep Revi...

引用

4th IEEE International Conference on Advance Computing and Innovative technologies in engineering, ICACITE 2024

作者： Aadithyan Shnain, Ammar Hameed Sharma, Vikrant Lafta, Alaa M. Mudhafar, Mustafa Ghobash, Ali Jawad, Ahmed Qusay Hindustan Institute of Technology and Science Department of Computer Applications India The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq Department of Cse Graphic Era Hill University Uttarakhand Dehradun248007 India Graphic Era Deemed to Be University Uttarakhand Dehradun248007 India National University of Science and Technology College of Technical Engineering Dhi Qar64001 Iraq University of Kerbala Faculty of Mediscal Applied Sciences Department of Medical Physics Karbala56001 Iraq Al-kunooze University College Computer Engineer Department Basra Iraq Kut University College Departmen of Electrical Engineering Technologies Wasit Iraq

ISBN: (纸本)9798350360165

Artificial intelligence (AI) is the process by which human-designed systems analyse available data and make judgements based on predetermined criteria in order to accomplish particular objectives. The incorporation of artificial intelligence (AI) has sparked a deep ethical conversation among key players, including tech companies, businesses, educational institutions, and legislators. AI's ethical problems have prompted the creation of ethical frameworks that aim to control its application for the good of society, the defence of human rights, and the maintenance of individual autonomy and privacy. This paper provides a critical evaluation and summary of the fundamental concepts behind the use of artificial intelligence, with a focus on the GDPR's (General Data Protection Regulation) approach. The rights of data subjects, data protection legislation, and data subjects' permission are crucial considerations in the context of 'privacy by design.' © 2024 IEEE.

关键词： Differential privacy

来源：评论

学校读者我要写书评

暂无评论

PROSOCIALDIALOG: A Prosocial Backbone for Conversational Agents

PROSOCIALDIALOG: A Prosocial Backbone for Conversational Age...

引用

2022 Conference on Empirical Methods in Natural language Processing, EMNLP 2022

作者： Kim, Hyunwoo Yu, Youngjae Jiang, Liwei Lu, Ximing Khashabi, Daniel Kim, Gunhee Choi, Yejin Sap, Maarten Allen Institute for Artificial Intelligence Department of Computer Science and Engineering Seoul National University Korea Republic of Paul G. Allen School of Computer Science University of Washington United States Johns Hopkins University United States Language Technologies Institute Carnegie Mellon University United States

Most existing dialogue systems fail to respond properly to potentially unsafe user utterances by either ignoring or passively agreeing with them. To address this issue, we introduce PROSOCIALDIALOG, the first large-scale multi-turn dialogue dataset to teach conversational agents to respond to problematic content following social norms. Covering diverse unethical, problematic, biased, and toxic situations, PROSOCIALDIALOG contains responses that encourage prosocial behavior, grounded in commonsense social rules (i.e., rules-of-thumb, RoTs). Created via a human-AI collaborative framework, PROSOCIALDIALOG consists of 58K dialogues, with 331K utterances, 160K unique RoTs, and 497K dialogue safety labels accompanied by free-form rationales. With this dataset, we introduce a dialogue safety detection module, Canary, capable of generating RoTs given conversational context, and a socially-informed dialogue agent, Prost. Empirical results show that Prost generates more socially acceptable dialogues compared to other state-of-the-art language and dialogue models in both in-domain and out-of-domain settings. Additionally, Canary effectively guides off-the-shelf language models to generate significantly more prosocial responses. Our work highlights the promise and importance of creating and steering conversational AI to be socially responsible. © 2022 Association for Computational Linguistics.

关键词： Large dataset

来源：评论

学校读者我要写书评

暂无评论

An Empirical Analysis of Various Techniques of Solving Obstacles through Artificial Intelligence 4

An Empirical Analysis of Various Techniques of Solving Obsta...

引用

4th IEEE International Conference on Advance Computing and Innovative technologies in engineering, ICACITE 2024

作者： Vaidya, Himadri Dhabliya, Dharmesh Jweeg, Muhseen Almusawi, Muntather Naser, Zamen Latef Hashem, Amjad Jawad, Ahmed Qusay Graphic Era Hill University Department of Cse India Graphic Era Deemed to Be University Uttarakhand Dehradun248007 India Information Technology Vishwakarma Institute of Information Technology Pune India Al-Farahidi University Baghdad Iraq The Islamic University College of Technical Engineering Department of Computers Techniques Engineering Najaf Iraq National University of Science and Technology College of Technical Engineering Dhi Qar64001 Iraq Alkunooze University College Computer Engineer Department Basra Iraq Kut University College Department of Electrical Engineering Technologies Wasit Iraq

ISBN: (纸本)9798350360165

In the space sector, artificial intelligence (AI), and especially machine learning (ML), is becoming more and more popular. These days, ML algorithms are extensively used in many different contexts, including operational control of satellite constellations, autonomous navigation, and spacecraft health monitoring. There are two main types of surveys that are now available on AI applications in space missions, each with certain drawbacks. Generally speaking, the first category is out of date since it does not include more recent advancements like Deep Learning (DL) and bioinspired AI algorithms. On the other hand, the second group tends to be too detailed and concentrates only on certain AI approaches, which can put off those who are looking for a more comprehensive knowledge, especially those with technical expertise in other subjects. In order to rectify these deficiencies, we provide a succinct summary that provides a thorough understanding of the uses and difficulties of AI in the space sector. Our study covers classical AI approaches used in several space mission disciplines, emphasizing their usefulness in real-world applications rather than going into formal methodologies. We also investigate bioinspired AI algorithms and point out possible uses for them. Beyond technical obstacles, we recognize and talk about significant legal ramifications of using AI in space projects. We expand our evaluation to include developments in Blockchain and Internet of Things (IoT) technology in the consideration of future directions. Specifically, we highlight how AI and Blockchain are convergent and pinpoint three major application fields in the space community that stand to gain a great deal from this convergence. These areas include satellite communications, spacecraft data security, and the Internet of Things in Space (IoST). Our review strives to be a useful tool for anyone interested in learning more about the complex world of artificial intelligence applications in space b

关键词： Satellite communication systems

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：