检索结果-内蒙古大学图书馆

Underwater Image Quality Evaluation: A Comprehensive Review

IET Image Processing 2025年第1期19卷

作者： Shen, Mengjiao Yang, Miao Zhong, Jinyang Liu, Hantao Pan, Can School of Electronic Engineering Jiangsu Ocean University Lianyungang China School of Computer Science and Informatics Cardiff University Cardiff United Kingdom

Underwater image quality evaluation (UIQE) is crucial in improving image processing techniques and optimizing the design of the imaging system to obtain object information more accurately. However, existing UIQE methods are designed based on limited images or consider only a few natural scene statistics (NSS) metrics, lacking consideration for generalization across various underwater imaging applications. In this paper, an in-depth review of the existing UIQE methods based on evaluation operations is provided, emphasizing the bias present when evaluating UIQE methods using individual metrics. To address this, a novel metric called quadrilateral datum evaluation (QDE) is designed for UIQE methods. It comprehensively considers robustness across different datasets, as well as correlation and ranking consistency with mean opinion scores (MOS). This is the first solution to measure an UIQE method from an all-encompassing visual perspective. By using QDE, UIQE methods characterized by greater feature strength and small imbalance demonstrate good consistency and robustness across multiple aspects, providing a basis for the design of UIQE methods. © 2025 The Author(s). IET Image Processing published by John Wiley & Sons Ltd on behalf of The Institution of engineering and Technology.

关键词： Image correlation

来源：评论

学校读者我要写书评

暂无评论

A Review of Link Prediction on Heterogeneous Networks 8th

A Review of Link Prediction on Heterogeneous Networks

引用

8th International Conference on Cognitive Computing, ICCC 2024, Held as Part of the Services Conference Federation, SCF 2024

作者： Mi, Rui Wang, Changbo Zhang, Liang-Jie Wu, Yirui Chen, Junyang Wang, Huan College of Informatics Huazhong Agricultural University Wuhan China Center of Health Administration and Development Studies Hubei University of Medicine Shiyan China College of Computer Science and Software Engineering Shenzhen University Shenzhen China Hohai University Nanjing China

ISBN: (纸本)9783031779534

Link prediction is a central downstream assignment in network analysis, which denotes an attempt to assess the probability of a connection between two nodes based on observed link and node properties, which can be utilized across diverse categories of graph system, including knowledge graphs, network traffic, social networks, biological networks, etc. Heterogeneous networks usually contain diverse nodes and a variety of edges’ type, which can preserve more information than homogeneous information networks. However, the majority of researches focus on graph representation analysis of homogeneous networks. Limited research has systematically reviewed advanced techniques for link prediction and applications on Heterogeneous Information Networks (HIN). Building upon this, we propose a classification summary for link prediction methods in heterogeneous information graphs. We will introduce some classical methods and some newly proposed methods for link prediction on heterogeneous graphs. We provide general technical ideas behind each method in classes and analysis performance of models. We further discuss the applications of link prediction on complex heterogeneous graphs across various domains. Additionally, the research challenges are discussed, and several possible research avenues for link prediction on heterogeneous network are pointed out. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Practical big data techniques for end-to-end machine learning deployment: a comprehensive review

引用

Discover Data 2025年第1期3卷 1-18页

作者： El-Sayed, Ahmed Abougabal, Mohamed Lazem, Shaimaa Computer and Systems Engineering Department Faculty of Engineering Alexandria University Alexandria Egypt Informatics Research Institute City of Scientific Research and Technological Applications New Borg El-Arab Egypt

In recent years, data generated from diverse sources has grown exponentially, giving rise to new challenges for processing and analysis under the umbrella of “big data". Although various methods and platforms have been proposed to tackle these issues, there is still a need for a conceptual guide that synthesizes state-of-the-art tools and techniques across the full machine learning (ML) workflow. In this survey, we consolidate research on big data infrastructures, distributed processing frameworks, and ML methods, mapping out an end-to-end conceptual framework that can serve as a reference architecture for end-to-end deployment. Specifically, we discuss key aspects of big data analytics (storage solutions, platforms like Hadoop and Spark, NoSQL databases), preprocessing approaches (dimensionality reduction, instance selection, noise handling), ML algorithms (supervised, unsupervised, and emerging deep learning paradigms), and deployment considerations (monitoring, continuous integration, and versioning). By gathering and integrating these elements, our survey provides a comprehensive overview of existing solutions, clarifies the design choices developers must consider, and identifies research gaps that remain to be addressed.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Improvement of the Teaching Process Using the Genetic Algorithm 8th

Improvement of the Teaching Process Using the Genetic Algori...

引用

8th EAI International Conference on Future Access Enablers of Ubiquitous and Intelligent Infrastructures, FABULOUS 2024

作者： Šimić, Goran Jevremović, Aleksandar Strugarević, Danilo School of Electrical and Computer Engineering at Academy of Technical and Art Applied Studies Belgrade11000 Serbia Faculty of Informatics and Computing Singidunum University Belgrade11000 Serbia Academy of Applied Preschool Teaching and Health Studies Kruševac37000 Serbia

ISBN: (纸本)9783031723926

Teaching is a process that requires permanent observation and improvement. With the rapid development of e-learning, there was a need to review, improve and optimize the process of evaluating students’ performance. The main objective of this study is to develop and implement the procedure for automatized generation of assessment tests. For this purpose, the proposed solution includes the application of a genetic algorithm. The procedure is developed on a sample of 29 existing test cases of "Internet programming" course. The optimization resulted in better selection and reduced number of questions used for evaluation of students` performance. The results of the study and the topicality of the issue point to the need for further research, both in the aspect of students` performance assessment, as well as in other aspects of higher education. © ICST Institute for computer Sciences, Social informatics and Telecommunications engineering 2025.

关键词： Conceptual framework e-learning Genetic algorithm Higher education Students` performance

来源：评论

学校读者我要写书评

暂无评论

Deep Learning for Cyber Security: Deephoax Detection 2

Deep Learning for Cyber Security: Deephoax Detection

引用

2nd International Conference on Advanced Innovations in Smart Cities, ICAISC 2025

作者： Yakub, Sy. Yuliani Wowor, Rivo Juicer Agustriawan, David Irmawati Kristiyanti, Dinar Ajeng Winarno, P.M. Hassan, Rosilah Mohamad, Roshayu Razzaq, Randy Zahran Universitas Multimedia Nusantara Department of Informatics Tangerang Indonesia Information System Study Program Universitas Multimedia Nusantara Tangerang Indonesia Universiti Kebangsaan Malaysia Selangor Malaysia College of Computer Science and Engineering University of Jeddah Jeddah Saudi Arabia Uin Sultan Maulana Hasanuddin Department of Informatics Banten Indonesia

ISBN: (纸本)9798331506995

The increasing prevalence of deep hoaxes, such as fake news and phishing schemes, poses a significant threat to cybersecurity, undermining trust and spreading misinformation. In Indonesia, surveys indicate that more than 60% of people exposed to hoax news believe that it is true, emphasizing the urgent need for robust detection methods. Traditional cybersecurity approaches often struggle to keep pace with the growing scale and sophistication of these attacks. To address this challenge, this research investigates the use of deep learning techniques, specifically focusing on text-based hoax detection in the Indonesian language. The study fine-tunes IndoBERT, a pretrained deep learning model optimized for Indonesian text, to enhance the accuracy and scalability of hoax detection. The IndoBERT model was trained on a balanced dataset of 29,552 articles, comprising both hoax and real news content, collected from the Mafindo API and Kaggle's Indonesia News Dataset. The model was fine-tuned using supervised learning and evaluated using several key metrics, including accuracy, F1-score, precision, and recall. The results demonstrate that IndoBERT outperforms existing state-of-the-art approaches, achieving an accuracy of 98.51%, an F1-score of 98.44%, and a precision of 98.23% on the test set. These results highlight the effectiveness of IndoBERT for hoax detection, which offers a scalable solution to improve cybersecurity defenses against deceptive content. This research contributes to the integration of advanced deep learning models into cybersecurity systems, addressing the evolving landscape of cyber threats. © 2025 IEEE.

关键词： Cyber attacks

来源：评论

学校读者我要写书评

暂无评论

End-to-End Optimized Ensemble Model for Alzheimer's Disease Detection

End-to-End Optimized Ensemble Model for Alzheimer's Disease ...

引用

International Conference on Ubiquitous Information Management and Communication (IMCOM)

作者： Abdelrahman Hassan Tamer Abuhmed Shaker El-Sappagh Faculty of Computer Science and Engineering Galala University Egypt College of Computing and Informatics Sungkyunkwan University South Korea

ISBN: (数字)9798331507817

ISBN: (纸本)9798331507824

Alzheimer's disease (AD) is a progressive neurodegenerative disorder with an increasing prevalence among the elderly, making early and accurate diagnosis critical for effective intervention and management. This paper introduces an end-to-end machine learning pipeline optimized for detecting AD using multimodal data from the Alzheimer's Disease Neuroimaging Initiative (ADNI). The study proposes an ensemble model that integrates key modalities-demographics, cognitive scores, neuropsychological assessments, and MRI data-via early fusion to enhance AD detection. Various machine learning models, including Logistic Regression, Gradient Boosting, AdaBoost, XGBoost, Support Vector Machine, and Neural Networks, are compared, and feature selection is optimized through Recursive Feature Elimination, ANOVA F-value, and mutual information methods, with mutual information showing the most efficacy. With further hyperparameter tuning, the highest accuracy is achieved at 95.85%, with Gradient Boosting as the top-performing model. Additional experiments assess and evaluate individual modality contributions, underscoring the predictive value of cognitive scores and neuropsychological features. To enhance clinical interpretability, we incorporate model explainability using LIME and SHAP, which identify key features influencing model predictions across different AD stages. Explainable AI (XAI) is essential in healthcare for providing transparency and supporting clinician confidence by highlighting which biomarkers, such as MMSE, CDRSB, and PTAU, contribute most to diagnostic outcomes. This study demonstrates classical machine learning's potential in AD diagnostics, offering a pathway to cost-effective and clinically viable models for neurodegenerative disease management.

关键词： Support vector machines Accuracy Biological system modeling Neural networks Predictive models Boosting Feature extraction Alzheimer's disease Tuning Optimization

来源：评论

学校读者我要写书评

暂无评论

3D Reconstruction of Colonic Polyps: A Methodology for Improved Visualization and Analysis

3D Reconstruction of Colonic Polyps: A Methodology for Impro...

引用

International Conference on Electrical, computer and Communication engineering (ECCE)

作者： Ishtiaque Ahmed Pranto Saha Md. Saimim Islam Khan Hamim Irfan Ahmad Alvi Md. Zahid Hasan Md. Hasan Imam Bijoy Department of Computer Science and Engineering Health Informatics Research Laboratory (HIRL) Daffodil International University Dhaka Bangladesh

ISBN: (数字)9798350357509

ISBN: (纸本)9798350357516

Polyps are critical abnormalities that may indicate the initial phases of colon cancer, making accurate detection and analysis essential in medical diagnostics. Traditional 2D medical imaging techniques provide limited insight into the shape and structure of polyps, which is where 3D reconstruction can play a transformative role. By converting 2D polyp images into 3D models, healthcare professionals can gain enhanced visualization, enabling better diagnosis, treatment planning, and surgical precision. This study demonstrates a robust process for reconstructing polyp images in 3D using the Kvasir-Seg dataset, which includes 1,000 polyp images with expertly annotated ground truth masks. Our methodology includes the generation of depth maps using the Intel DPT-Large model, the extraction of point clouds with a customized approach, and the construction of precise mesh objects with Open3D. We extract geometric properties, including linearity, planarity, and curvature change, from the point clouds and meshes for a more in-depth analysis of the 3D objects. The unsupervised analysis of the resulting density plots indicates the presence of two distinct clusters within the feature distributions, highlighting the significance of 3D reconstruction for improved polyp characterization. Furthermore, the evaluation included 2D features such as edge density, contrast, dissimilarity, homogeneity, energy, correlation, entropy, number of lines, circularity, and Gabor energy. The analysis of p-values reveals that the 3D features related to these two clusters show notably lower p-values, thereby demonstrating their significance in the comprehensive analysis.

关键词： Point cloud compression Visualization Solid modeling Three-dimensional displays Accuracy Shape Surgery Transforms Planning Image reconstruction

来源：评论

学校读者我要写书评

暂无评论

Comprehensive Classification of Web Tracking Systems: Technological Insights and Analysis

arXiv

引用

arXiv 2025年

作者： Tasoulas, Theofanis Gazis, Alexandros Tsohou, Aggeliki Ionian University Department of Informatics Corfu Greece Democritus University of Thrace Department of Electrical Engineering and Computer Greece

Web tracking (WT) systems are advanced technologies used to monitor and analyze online user behavior. Initially focused on HTML and static webpages, these systems have evolved with the proliferation of IoT, edge computing, and Big Data, encompassing a broad array of interconnected devices with APIs, interfaces and computing nodes for interaction. WT systems are pivotal in technological innovation and business development, although trends like GDPR complicate data extraction and mandate transparency. Specifically, this study examines WT systems purely from a technological perspective, excluding organizational and privacy implications. A novel classification scheme based on technological architecture and principles is proposed, compared to two preexisting frameworks. The scheme categorizes WT systems into six classes, emphasizing technological mechanisms such as HTTP protocols, APIs, and user identification techniques. Additionally, a survey of over 1,000 internet users, conducted via Google Forms, explores user awareness of WT systems. Findings indicate that knowledge of WT technologies is largely unrelated to demographic factors such as age or gender but is strongly influenced by a user's background in computer science. Most users demonstrate only a basic understanding of WT tools, and this awareness does not correlate with heightened concerns about data misuse. As such, the research highlights gaps in user education about WT technologies and underscores the need for a deeper examination of their technical underpinnings. This study provides a foundation for further exploration of WT systems from multiple perspectives, contributing to advancements in classification, implementation, and user awareness. © 2025, CC BY.

关键词： HTTP

来源：评论

学校读者我要写书评

暂无评论

LATUP-Net: A lightweight 3D attention U-Net with parallel convolutions for brain tumor segmentation

引用

computers in Biology and Medicine 2025年 184卷 109353-109353页

作者： Alwadee, Ebtihal J. Sun, Xianfang Qin, Yipeng Langbein, Frank C. School of Computer Science and Informatics Cardiff University CardiffCF24 4AG United Kingdom Department of Computer Science College of Engineering and Computer Science Jazan University Jazan Saudi Arabia

Early-stage 3D brain tumor segmentation from magnetic resonance imaging (MRI) scans is crucial for prompt and effective treatment. However, this process faces the challenge of precise delineation due to the tumors’ complex heterogeneity. Moreover, energy sustainability targets and resource limitations, especially in developing countries, require efficient and accessible medical imaging solutions. The proposed architecture, a Lightweight 3D ATtention U-Net with Parallel convolutions, LATUP-Net, addresses these issues. It is specifically designed to reduce computational requirements significantly while maintaining high segmentation performance. By incorporating parallel convolutions, it enhances feature representation by capturing multi-scale information. It further integrates an attention mechanism to refine segmentation through selective feature recalibration. LATUP-Net achieves promising segmentation performance: the average Dice scores for the whole tumor, tumor core, and enhancing tumor on the BraTS 2020 dataset are 88.41%, 83.82%, and 73.67%, and on the BraTS 2021 dataset, they are 90.29%, 89.54%, and 83.92%, respectively. Hausdorff distance metrics further indicate its improved ability to delineate tumor boundaries. With its significantly reduced computational demand using only 3.07 M parameters, about 59 times fewer than other state-of-the-art models, and running on a single NVIDIA GeForce RTX3060 12 GB GPU, LATUP-Net requires just 15.79 GFLOPs. This makes it a promising solution for real-world clinical applications, particularly in settings with limited resources. Investigations into the model's interpretability, utilizing gradient-weighted class activation mapping and confusion matrices, reveal that while attention mechanisms enhance the segmentation of small regions, their impact is nuanced. Achieving the most accurate tumor delineation requires carefully balancing local and global features. The code is available at https://***/ca/code-bca. © 2024 T

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Transforming Synthesized Netlists into a Feasible & Compliant Functional Bookshelf Format 24

Transforming Synthesized Netlists into a Feasible & Complian...

引用

Proceedings of the 28th Pan-Hellenic Conference on Progress in Computing and informatics

作者： George Kranas Kosmas Kariotis Antonios Dadaliaris George Stamoulis Athanasios Kakarountas Dept. of Computer Science & Biomedical Informatics University of Thessaly Lamia Greece Dept. of Informatics & Telecommunications University of Thessaly Lamia Greece Dept. of Electrical and Computer Engineering University of Thessaly Volos Greece

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：