检索结果-内蒙古大学图书馆

2nd International Conference on Pervasive Computing Advances and Applications, PerCAA 2024

作者： Swarnkar, Shekhar Kumar Yadav, Surendra Kumar Dept. of Computer Science & Engineering JECRC University Rajasthan Jaipur India Dept. of Computer Science & Engineering JECRC University Rajasthan Jaipur India

ISBN: (纸本)9781837242627

Under the brain tumor situation major poses risk to human life, and the timely detection of this illness is to successfully save human life. The suitable conclusion about brain tumor is the practice of using precise and efficient approaches for segmentation and classification on Magnetic Resonance Imaging. Major categorization of aberrant and normal tumor MR images has been the focus of research that has already been conducted. When it comes to classifying brain tumor cells, this paper presents a relative examination of various machine learning approaches that are considered to be state-of-the-art. The paper present both supervised and unsupervised learning methods, such as Support Vector Machines (SVMs), K-Nearest Neighbors (KNNs), Random Forests (RF), and deep learning models. A comparative analysis are presented in this research that focus on the tactics that are now being used that have shown to be the most successful in their brain tumor detection. There is a wide range of classifiers that may be used to get feature extraction from MRI brain images. In order to define the abnormal developments in brain MRI scans, the main phase of these procedures comprises a few actions that must be carried out while applying different techniques. These actions must be carried out in order to define the abnormal developments. A variety of models that are associated with feature extraction framework are used by the majority of the approaches that are now in use to extract substantial information from MRI brain images. © The Institution of engineering & Technology 2024.

关键词： Artificial Intelligence Brain Tumor Machine Learning Prediction support Vector Machine

来源：评论

学校读者我要写书评

暂无评论

Image-Driven Face Presentation Attack Detection in Biometric Systems

Image-Driven Face Presentation Attack Detection in Biometric...

引用

2025 International Conference on Emerging Technologies for Intelligent Systems, ETIS 2025

作者： Vijayan, Rini Salim, A. College of Engineering Trivandrum Affliated to Apj Abdul Kalam Technological University Dept. of Computer Science and Engineering Kerala India

ISBN: (纸本)9798331507541

As biometric systems become increasingly relied on secure authentication, concerns about their susceptibility to Face Presentation Attacks (FPAs) have intensified. This paper introduces a framework for detection of image-driven FPAs in biometric systems, focusing on enhancing the security of face recognition technologies. We propose a methodology that integrates Histogram of Oriented Gradients (HOG) with deep learning techniques. This approach utilizes pretrained models and custom layers to effectively differentiate between legitimate user images and various forms of presentation attacks, including those generated by GANs, morphing techniques, and photoshop manipulations. Through extensive experimentation, we analyzed the performance of each model, offering detailed comparisons concerning accuracy, precision, and F1 score across various categories as well as overall performance of each model. Our findings imply that fusion of HOG with a fine-tuned VGG16 model can achieve significantly higher overall accuracy. © 2025 IEEE.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Fig Leaf Disease Prediction and Severity Estimation UsingDeep Learning

Fig Leaf Disease Prediction and Severity Estimation UsingDee...

引用

2024 International Conference on Computing and Intelligent Reality Technologies, ICCIRT 2024

作者： Prasanna Kumar, K.R. Gunasekar, M. Logeswaran, K. Dhanya Sree, P. Malathi, P. Vignesh, V. Dept. of Computer Science and Design Kongu Engineering College Tamil Nadu Erode India School of Computer Science and Engineering VIT Tamil Nadu Vellore India Dept. of Artificial Intelligence Kongu Engineering College Tamil Nadu Erode India

ISBN: (纸本)9798331510299

Fig leaf diseases are among the major causes of crop loss. Better methods for detection and further increased accuracy in the estimation of the severity of the disease are demanded. In basis of what is previously researched, this project continues with not just the identification of the disease but also the identification of the level of infections. Having a large dataset of 1, 871 images that comprises 971 healthy samples and 900 infected samples, the analysis will be more elaborate to determine the disease in comparison with previous estimates. We have used two deep learning models - GoogLeNet and AlexNet with outstanding performance in preceding research to make reliable predictions on leaf diseases. The novel contribution of our work is estimating the disease severity with the help of Convolutional Neural Networks that could extract features and classify disease. For the estimation of the disease severity, we will use the VGG16 model to be able to categorize the infection severity into three classes- healthy, mild, and high. The diseases addressed in this paper are Cerotelium fici (fig rust) and Fig mosaic *** goal is to offer a comprehensive tool designed to identify diseases in fig crops and assess their severity for fig farmers and the researcher so that timely disease management strategies could be undertaken. This integrated approach not only aids in taking prompt intervention but also contributes significantly toward sustainable agriculture by reducing adverse impacts on fig leaf diseases. © 2024 IEEE.

关键词： AlexNet Deep Learning GoogLeNet Plant Disease prediction Severity Estimation VGG16

来源：评论

学校读者我要写书评

暂无评论

Chain-of-Factors: A Zero-Shot Prompting Methodology Enabling Factor-Centric Reasoning in Large Language Models 23

Chain-of-Factors: A Zero-Shot Prompting Methodology Enabling...

引用

23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024

作者： Hussain, Musarrat Ur Rehman, Ubaid Nguyen, Tri D.T. Lee, Sungyoung Kim, Seong Tae Bae, Sung-Ho Kim, Jung Uk UiT The Arctic University of Norway Department of Computer Science Tromso Norway Kyung Hee University Dept. of Computer Science & Engineering Yongin Korea Republic of Kyung Hee University Department of Computer Science and Engineering Yongin Korea Republic of

ISBN: (纸本)9798350374889

Large language models (LLMs) have significantly improved numerous natural language processing tasks. However, their performance relies heavily on the provided instructions or prompts. Recently, several prompting methodologies have been developed to enhance the reasoning abilities of LLMs. Notably, the Chain-of-Thought (CoT) approach provides examples that help break down tasks into sub-steps, resulting in more accurate solutions. However, the process of generating detailed examples may not be user-friendly, as end users prefer providing task descriptions rather than a set of examples. In this study, we introduce Chain-of-Factors (CoF), an innovative zero-shot prompting methodology that incorporates task-specific instructions as a chain of factors into the prompt, aimed at enhancing the factor-centric reasoning abilities of LLMs. Experiments on three LLMs, including ChatGPT-3.5, Gemini, and GPT-4, show performance improvements ranging from 0.01% to 40.2% in accuracy on various symbolic reasoning and logical reasoning tasks compared with zero-shot and few-shot CoT. In summary, CoF enhances LLMs' reasoning abilities by including task-specific steps and instructions, while also decreasing the necessity for fine-tuning specific to each task. © 2024 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

A Medium Voltage Grid-connected PV Inverter with a New Modular High Voltage Gain Converter Featuring Internal Modified Voltage Doubling Balancers 14

A Medium Voltage Grid-connected PV Inverter with a New Modul...

引用

14th Annual IEEE Applied Power Electronics Conference and Exposition, APEC 2025

作者： Kanathipan, Kajanan Cheema, Muhammad Ali Masood Lam, John York University Dept. of Electrical Engineering and Computer Science Toronto Canada Northern Transformer Toronto Canada

ISBN: (纸本)9798331516116

This work proposes a medium voltage grid-connected inverter with modular high voltage gain converters for PV energy applications. The proposed topology utilizes (1) PV arrays interfaced with modular DC-DC converters in an input independent output series (IIOS) configuration, (2) internal modified voltage doubling balancers to achieve balanced output voltage under power mismatch scenarios, and (3) a high voltage neutral point clamped (NPC) 5-level grid-connected inverter system. The presented DC-DC converters are able to perform maximum power point tracking operation through phase-shift control on their full-bridge switches and ensure balanced output voltage under all scenarios with duty ratio control on the internal voltage doublers. In addition, a high step-up gain and soft-switching operation on all module switches are achieved through the use of an LLC resonant stage. A five-level three-phase inverter interfaces the system with the MVAC grid. The steady-state and dynamic performance of the proposed modular system are validated through simulation results on a 27kVrms (L-L) 60Hz output, four-module 60kW system and preliminary experimental results on a two-module proof-of-concept 1kV DC-link laboratory prototype. © 2025 IEEE.

关键词： Electric power transmission networks

来源：评论

学校读者我要写书评

暂无评论

A Study on the Efficient ESS Charging/Discharging Operation Algorithm in LVDC Microgrid Environment

引用

Transactions of the Korean Institute of Electrical Engineers 2024年第6期73卷 1063-1073页

作者： Jung, Sang-Woo An, Yoon-Young Kim, Ki-Il Dept. of Computer Science and Engineering Chungnam National University Korea Republic of ICT Convergence Standards Research Division Electronics and Telecommunications Research Institute Korea Republic of

Modern power systems are developing in a sustainable and energy-efficient direction, and research on intelligent power grids that accommodate new and renewable energy is being actively conducted. Since new and renewable energy-based distributed power generation systems and ESS are output as DC power, the DC microgrid reduces the power conversion step and improves the stability and efficienc of the power distribution network. The core of this study is to compare and analyze the performance of the three charge/discharge operation algorithms based on AI prediction in the LVDC microgrid environment. The AI server learns real-time data through interworking with the EMS to predict future load usage and solar power generation, and performs the charge/discharge operation algorithm of ESS based on the collected ESS state data. This paper presents the configuration of the LVDC microgrid testbed, the integration method between EMS and electrical facilities, and analyzes the performance of the developed ESS charge/discharge operation algorithm. It also evaluates the effectiveness of each operation algorithm through actual operation data and simulation, analyzes how these strategies contribute to system load reduction and energy efficiency improvement, and provides guidelines for selecting the optimal ESS charge/discharge operation algorithm under specific conditions. © The Korean Institute of Electrical Engineers.

关键词： Solar energy

来源：评论

学校读者我要写书评

暂无评论

A Novel Framework for Real-Time Fire Detection in CCTV Videos Using a Hybrid Approach of Motion-Flicker Detection, Colour Detection and YOLOv7

A Novel Framework for Real-Time Fire Detection in CCTV Video...

引用

2023 International Conference on Electrical, computer, Communications and Mechatronics engineering, ICECCME 2023

作者： Jamal, Sangeetha Jackson, Henok Benny, Evin Koshy, Jeroen Dani Joseph, Gloria School of Engineering and Technology Dept. of Computer Science and Engineering Rajagiri Kochi India

ISBN: (纸本)9798350322972

This paper provides a practical view of implementing a real-time vision-based fire detection system that can accurately detect and alert the presence of fires in live CCTV footage. The proposed system combines motion-flicker detection, which uses a background subtraction algorithm to extract the moving objects in a video and a flicker detection algorithm to verify flickering in suspected regions. Then using a YCbCr colour model, the colour of the flickering region is verified to match the characteristics of fire. Further, a YOLOv7 model trained on a large dataset of fire images is employed to verify the results and raises the alarm to alert the presence of fire. Several optimizations have been done to reduce false alarm rate and GPU memory usage times. The accuracy is 98.4%, and false alarm rate is as low as 2.1%. © 2023 IEEE.

关键词： Fires

来源：评论

学校读者我要写书评

暂无评论

Source Free Domain Adaptation with Pseudo-Labeling Quality Assessed by Sam in Fundus Image Segmentation 22

Source Free Domain Adaptation with Pseudo-Labeling Quality A...

引用

22nd IEEE International Symposium on Biomedical Imaging, ISBI 2025

作者： Yaacovi, Bar Goldberger, Jacob Tel-Aviv University Computer Science Dept. Israel Bar-Ilan University Faculty of Engineering Israel

ISBN: (纸本)9798331520526

Precise segmentation of the optic cup and optic disc in fundus images is crucial for measuring the cup-to-disc ratio, a key indicator for glaucoma screening and diagnosis. When applying a network trained on data from one medical site to data from another site, performance declines due to domain shift problems. This study centers on the realistic scenario of source-free unsupervised domain adaptation. We propose a new pseudo-labeling strategy that takes advantage of the capabilities of the SAM segmentation foundation model by using it to assess the quality of the pseudo-label segmentation of the target images created by the source model. We implemented our method on publicly available retinal segmentation datasets collected from several medical sites and reported improved performance compared to previous methods. © 2025 IEEE.

关键词： Image segmentation

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Based Estimation of Dynamic Travel Times Between Points of Interest

Machine Learning-Based Estimation of Dynamic Travel Times Be...

引用

2025 IEEE International Conference on Consumer Electronics, ICCE 2025

作者： Kaji, Shoma Kajimoto, Dai Noguchi, Tatsuya Takayama, Toshinori Bao, Siya Togawa, Nozomu Dept. of Computer Science and Communications Engineering Waseda University Tokyo Japan ZENRIN DataCom Tokyo Japan

ISBN: (纸本)9798331521165

Due to the dynamic traffic conditions, effective estimation of travel times is crucial for the real-world travel planning service. Although APIs provide accurate travel information, the high response time poses a limitation. This paper proposes a machine learning-based method using XGBoost to predict dynamic travel times between Points of Interests (POIs). By utilizing features including latitude and longitude of locations, route direction, and departure time, the model achieves high prediction accuracy. We conducted hyper-parameter tuning using Optuna and evaluated the proposed method with POIs in Kyoto. The proposed method achieved a low RMSPE of −3s. © 2025 IEEE.

关键词： Travel time

来源：评论

学校读者我要写书评

暂无评论

Autism Spectrum Disorder Detection Using Convolutional Neural Network

Autism Spectrum Disorder Detection Using Convolutional Neura...

引用

2025 International Conference on Visual Analytics and data Visualization, ICVADV 2025

作者： Petricia, V. Angel Jacksy, M. Bhavadharani, R. St. Joseph's Institute of Technology Dept. of Computer Science and Engineering Chennai119 India

ISBN: (纸本)9798331521394

Autism Spectrum Disorder(ASD) is a complex neurodevelopmental condition marked by challenges in social interaction, communication, and repetitive behaviors. Early detection is critical to enable timely intervention and improve outcomes for individuals with ASD. This research proposes an innovative framework that integrates Convolutional Neural Networks (CNNs) with advanced data preprocessing techniques, facial imaging, and behavioral analysis to enhance the accuracy and efficiency of ASD diagnosis. Leveraging data augmentation and transfer learning, the model achieves robust performance with high sensitivity and specificity. The novelty of this approach lies in its ability to identify subtle facial and behavioral patterns indicative of ASD, using a multidisciplinary strategy that combines neuroimaging, machine learning, and real-time diagnostic feedback. By advancing beyond traditional methods, this comprehensive framework holds the potential to revolutionize early ASD detection, making it more accessible, scalable, and precise. © 2025 IEEE.

关键词： Convolutional neural networks

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：