检索结果-内蒙古大学图书馆

8th International Conference on Video and Image Processing, ICVIP 2024

作者： Em, Tithnorakneath Aramvith, Supavadee Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok Thailand Multimedia Data Analytics and Processing Research Unit Department of Electrical Engineering Faculty of Engineering Chulalongkorn University Bangkok Thailand

ISBN: (数字)9781510689244

ISBN: (纸本)9781510689237

Scene Text Image Super-Resolution (STISR) has become a significant pre-processing technique to enhance text recognition accuracy in low-resolution scene text images, particularly when addressing the challenges of low-resolution (LR) images in real-world applications. Despite advancements in general and text-focused methods, current techniques often fail to retain the fine-grained and text-specific details necessary for downstream recognition tasks. This paper presents a novel Attention-Guided Scene Text Image Super-Resolution (AG-STISR) framework designed to overcome these challenges. Our approach integrates advanced channel attention and spatial attention mechanisms within a shallow feature extraction module, enabling precise enhancement of key text features such as character edges, contours, and fine structural details. Channel attention helps prioritize the features rich in text-specific details by focusing the model's attention on the channels with the most important data for text recognition. On the other hand, Spatial attention emphasizes locating and highlighting the most significant spatial areas in the image, like character outlines and strokes, ensuring that key elements are maintained while minimizing background distractions. We extensively evaluated our model on the TextZoom dataset and assessed it against various scene text recognition standards. The results demonstrate that our AG-STISR model consistently outperforms state-of-the-art methods, showing significant improvements in PSNR, SSIM, and text recognition accuracy across varying levels of image degradation. This research introduces a compelling solution to STISR, enhancing not only image clarity but also the reliability of text recognition in real-world applications. © 2025 SPIE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

Normalized Orthography for Tunisian Arabic 3rd

Normalized Orthography for Tunisian Arabic

引用

3rd International Conference on Language Processing and Knowledge Management, LPKM 2024

作者： Turki, Houcemeddine Ellouze, Kawthar Ben Ammar, Hager Hadj Taieb, Mohamed Ali Adel, Imed Ben Aouicha, Mohamed Farri, Pier Luigi Bennour, Abderrezak Data Engineering and Semantics Research Unit University of Sfax Sfax Tunisia Association Derja Tunis Tunisia University of Sousse Sousse Tunisia VBScuola Rome Italy

ISBN: (纸本)9783031850660

Tunisian Arabic (ISO 693-3: aeb) is a distinct variety native to Tunisia, derived from Arabic and enriched by various historical influences. This research introduces the "Normalized Orthography for Tunisian Arabic" (NOTA), an adaptation of CODA* guidelines for transcribing Tunisian Arabic using Arabic script. The aim is to enhance language resource development by ensuring user-friendliness and consistency. The updated standard addresses challenges in accurately representing Tunisian phonology and morphology, correcting issues from transcriptions based on Modern Standard Arabic. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Arabic Script Normalization Orthography Tunisian

来源：评论

学校读者我要写书评

暂无评论

Resilience in Online Federated Learning: Mitigating Model-Poisoning Attacks via Partial Sharing

引用

IEEE Transactions on Signal and Information Processing over Networks 2025年 11卷 388-400页

作者： Lari, Ehsan Arablouei, Reza Gogineni, Vinay Chakravarthi Werner, Stefan Commonwealth Scientific and Industrial Research Organisation PullenvaleQLD4069 Australia University of Southern Denmark Applied AI and Data Science Unit The Maersk Mc-Kinney Moller Institute Faculty of Engineering Odense5230 Denmark Norwegian University of Science and Technology Department of Electronic Systems Trondheim7034 Norway Aalto University Department of Information and Communications Engineering Espoo02150 Finland

Federated learning (FL) allows training machine learning models on distributed data without compromising privacy. However, FL is vulnerable to model-poisoning attacks where malicious clients tamper with their local models to manipulate the global model. In this work, we investigate the resilience of the partial-sharing online FL (PSO-Fed) algorithm against such attacks. PSO-Fed reduces communication overhead by allowing clients to share only a fraction of their model updates with the server. We demonstrate that this partial sharing mechanism has the added advantage of enhancing PSO-Fed's robustness to model-poisoning attacks. Through theoretical analysis, we show that PSO-Fed maintains convergence even under Byzantine attacks, where malicious clients inject noise into their updates. Furthermore, we derive a formula for PSO-Fed's mean square error, considering factors like stepsize, attack probability, and the number of malicious clients. Interestingly, we find a non-trivial optimal stepsize that maximizes PSO-Fed's resistance to these attacks. Extensive numerical experiments confirm our theoretical findings and showcase PSO-Fed's superior performance against model-poisoning attacks compared to other leading FL algorithms. © 2015 IEEE.

关键词： Federated learning

来源：评论

学校读者我要写书评

暂无评论

Performance Evaluation of Pond Ash-Enhanced Flowable Fill for Plastic Concrete Cutoff Walls in Earthen Dams Using Advanced Machine Learning Models

引用

Arabian Journal for Science and engineering 2025年 1-25页

作者： Dev, K. Lini Kumar, Divesh Ranjan Wipulanusat, Warit Department of Civil Engineering National Institute of Technology Patna India Research Unit in Data Science and Digital Transformation Department of Civil Engineering Faculty of Engineering Thammasat School of Engineering Thammasat University Pathumthani Thailand

Structural damage and foundation leakage are major concerns for earthen dams. To minimize seepage, cutoff walls are typically installed beneath the dam core to act as impermeable barriers. While concrete cutoff walls are widely used, their limited ductility and strength incompatibility with foundation soil present design challenges. Plastic concrete, a modified form of conventional concrete incorporating bentonite and pond ash, offers improved ductility and reduced brittleness, making it a suitable alternative. This study investigates the use of pond ash-based flowable fill as a replacement for normal concrete in plastic concrete cutoff walls. The unconfined compressive strength (UCS) of plastic concrete mixes was analyzed using four advanced regression machine learning algorithms: multivariate adaptive regression splines, extreme neural network (ENN), extreme gradient boosting (XGBoost), and gradient boosting machine (GBM). Several performance indices were used to evaluate model accuracy. The MARS model achieved the highest accuracy, with R² = 0.990 for training and R² = 0.963 for testing, followed by XGBoost, GBM, and ENN. SHAP analysis revealed that curing period has the most significant positive effect on UCS, followed by water and cement contents, while bentonite showed the least impact. Key properties were evaluated to determine an optimal mix design. This research enhances the understanding of CLSM-based plastic concrete and supports its application in cutoff walls by developing accurate UCS prediction models, contributing to the improved suitability and sustainability of dam foundation systems. © King Fahd University of Petroleum & Minerals 2025.

关键词： Compressive strength Cutoff walls Flowability Multivariate adaptive regression splines (MARS) Permeability Shear strength

来源：评论

学校读者我要写书评

暂无评论

Enhancing malware detection performance: leveraging K-Nearest Neighbors with Firefly Optimization Algorithm

引用

Multimedia Tools and Applications 2025年第12期84卷 10071-10094页

作者： Al Saaidah, Adeeb Abualhaj, Mosleh M. Shambour, Qusai Y. Abu-Shareha, Ahmad Adel Abualigah, Laith Al-Khatib, Sumaya N. Alraba’nah, Yousef H Department of Networks and Cybersecurity Al-Ahliyya Amman University Amman Jordan Department of Software Engineering Al-Ahliyya Amman University Amman Jordan Department of Data Science and Artificial Intelligence Al-Ahliyya Amman University Amman Jordan Hourani Center for Applied Scientific Research Al-Ahliyya Amman University Amman19328 Jordan Research Center University of Tabuk Tabuk71491 Saudi Arabia MEU Research Unit Middle East University Amman11831 Jordan Department of Electrical and Computer Engineering Lebanese American University Byblos13-5053 Lebanon College of Engineering Yuan Ze University Taoyuan32003 Taiwan Applied Science Research Center Applied Science Private University Amman11931 Jordan School of Engineering and Technology Sunway University Malaysia Petaling Jaya27500 Malaysia

Malware detection plays a crucial role in ensuring robust cybersecurity amidst the ever-evolving cyber threats. This research paper delves into the realm of machine learning (ML) algorithms for malware detection, with a specific emphasis on the K-Nearest Neighbors (KNN) algorithm, utilizing tailored parameter settings and the Firefly Optimization Algorithm (FOA). The study leverages the MalMem-2022 dataset to assess the efficacy of KNN and KNN with FOA in malware detection. The impact of parameter tuning and feature selection on classification is elucidated by comparing the performance of both approaches. Encouragingly, the results reveal promising advancements in one of the multiclass classification scenarios when employing KNN with FOA, yielding noteworthy enhancements in Accuracy, Recall, Precision, Matthews Correlation Coefficient, TNR, and F1-score by 2.65%, 2.65%, 2.47%, 3.59%, 0.17%, and 2.61%, respectively. These findings accentuate the significance of optimizing KNN’s parameters and implementing FOA for feature selection, culminating in heightened accuracy in malware detection. The symbiotic fusion of KNN and FOA exhibits remarkable effectiveness in augmenting the classification model’s performance, thus offering valuable insights for bolstering cybersecurity measures. Adopting ML algorithms, particularly KNN with specific parameter settings and FOA, presents a promising avenue for enhancing malware detection capabilities. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2024.

关键词： Nearest neighbor search

来源：评论

学校读者我要写书评

暂无评论

Text Categorization Can Enhance Domain-Agnostic Stopword Extraction 3rd

Text Categorization Can Enhance Domain-Agnostic Stopword Ext...

引用

3rd International Conference on Language Processing and Knowledge Management, LPKM 2024

作者： Turki, Houcemeddine Etori, Naome A. Hadj Taieb, Mohamed Ali Omotayo, Abdul-Hakeem Emezue, Chris Chinenye Ben Aouicha, Mohamed Awokoya, Ayodele Lawan, Falalu Ibrahim Nixdorf, Doreen Data Engineering and Semantics Research Unit University of Sfax Sfax Tunisia University of Minnesota MinneapolisMN United States University of California DavisCA United States Technical University of Munich Munich Germany McPherson University Lagos Nigeria Kaduna State University Kaduna Nigeria Masakhane Research Community Johannesburg South Africa

ISBN: (纸本)9783031850660

This paper investigates the role of text categorization in streamlining stopword extraction in natural language processing (NLP), specifically focusing on nine African languages alongside French. By leveraging the MasakhaNEWS, African Stopwords Project, and MasakhaPOS datasets, our findings emphasize that text categorization effectively filters out domain-specific keywords for most examined languages, allowing a simpler identification of stopwords. Nevertheless, linguistic variances result in lower detection rates for specific languages. Interestingly, we find that while over 40% of stopwords are common across news categories, less than 15% are unique to a single category. Uncommon stopwords add depth to the text, but their classification as stopwords depends on context. Therefore, combining statistical and linguistic approaches creates comprehensive stopword lists, highlighting the value of our hybrid method. This research enhances NLP for African languages and underscores the importance of text categorization in stopword extraction. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： African Languages Domain-agnostic Stopwords Stopword Extraction Text Categoization

来源：评论

学校读者我要写书评

暂无评论

Topological Analysis of Dermatoglyphic Patterns using Synthetic and Real data 25

Topological Analysis of Dermatoglyphic Patterns using Synthe...

引用

26th International Conference on Distributed Computing and Networking, ICDCN 2025

作者： Devkar, Prasad D. Ajwani, Nalini Malla, Sudarshan Bhatt, Aasttha Dave, Mahiman Sharma, Neha Panday, Mrityunjoy Yajnik, Chittaranjan S K. J. Somaiya Institute of Technology Sion Maharashtra Mumbai India Bhilai Institute of Technology Durg Chhattisgarh Bhilai India Pragati Engineering College Andhrapradesh Kakinada India Society for Data Science Maharashtra Pune India Society for Data Science West Bengal Kolkata India Diabetes Unit KEM Hospital & Research Center Maharashtra Pune India

ISBN: (纸本)9798400710629

This study presents an innovative approach for fingerprint extraction and analysis from hospital forms, using the Segment Anything Model (SAM) and Topological data Analysis (TDA). Synthetic data from the IISC Anguli dataset was first used to refine the methodology before applying it to real-world data from the Diabetes unit, KEM Hospital research Centre, Pune. Fingerprint segmentation using SAM isolates prints from complex backgrounds, with preprocessing steps improving clarity. Semantic sketches visually highlight critical dermatoglyphic details for feature identification. Minutiae points, such as ridge endings and bifurcations, are plotted to generate persistence barcodes, capturing essential fingerprint features. A distance matrix then evaluates similarities between fingerprints, enhancing biometric identification and verification. This research addresses technical challenges in medical data management, advancing biometric identification and improving healthcare data handling. © 2025 Copyright held by the owner/author(s).

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Accurate and Efficient Real-World Fall Detection Using Time Series Techniques 9th

Accurate and Efficient Real-World Fall Detection Using Time...

引用

9th International Workshop on Advanced Analytics and Learning on Temporal data, AALTD 2024

作者： Aderinola, Timilehin B. Palmerini, Luca D’Ascanio, Ilaria Chiari, Lorenzo Klenk, Jochen Becker, Clemens Caulfield, Brian Ifrim, Georgiana Insight SFI Research Centre for Data Analytics University College Dublin Dublin Ireland Department of Electrical Electronic and Information Engineering "Guglielmo Marconi" University of Bologna Bologna Italy Department of Clinical Gerontology Robert Bosch Hospital Stuttgart Germany Digital Geriatrics Unit Heidelberg University Hospital Heidelberg Germany

ISBN: (纸本)9783031770654

Falls pose a significant health risk, particularly for older people and those with specific medical conditions. Therefore, timely fall detection is crucial for preventing fall-related complications. Existing fall detection methods often have high false alarm or false negative rates, and many rely on handcrafted features. Additionally, most approaches are evaluated using simulated falls, leading to performance degradation in real-world scenarios. This paper explores a new fall detection approach leveraging real-world fall data and state-of-the-art time series techniques. The proposed method eliminates the need for manual feature engineering and has efficient runtime. Our approach achieves high accuracy, with false alarms and false negatives each as few as one in three days on FARSEEING, a large dataset of real-world falls (mean F1 score: 90.7%). We also outperform existing methods on simulated falls datasets, FallAllD and SisFall. Furthermore, we investigate the performance of models trained on simulated data and tested on real-world data. This research presents a real-time fall detection framework with potential for real-world implementation. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Fall detection

来源：评论

学校读者我要写书评

暂无评论

Tunisian Normalized Pronunciation 3rd

Tunisian Normalized Pronunciation

引用

3rd International Conference on Language Processing and Knowledge Management, LPKM 2024

作者： Turki, Houcemeddine Gibson, Maik Zribi, Rafik Adel, Imed Hadj Taieb, Mohamed Ali Ben Aouicha, Mohamed Ellouze, Kawthar Ben Ammar, Hager Bennour, Abderrezak Data Engineering and Semantics Research Unit Faculty of Sciences of Sfax University of Sfax Sfax Tunisia Redcliffe College Gloucester United Kingdom IÉSEG School of Management Lille France École supérieure des sciences et de la technologie de Hammam Sousse Université de Sousse Sousse Tunisia Concordia University Montréal Canada Lycée Gustave-Flaubert La Marsa Ariana Tunisia Institut Supérieur des Langues de Gabès Université de Gabès Gabès Tunisia

ISBN: (纸本)9783031850660

Tunisian (ISO 693-3: aeb) is a set of linguistic varieties spoken in Tunisia and evolved from the Arabic language. In this research work, we attempt to establish a standardized pronunciation that can serve as a communication tool for the diverse Tunisian population. This paper draws on prior linguistic research and uses a rich dataset maintained by the Linguistic data Consortium to identify common phonological features across various Tunisian dialects. This research creates a harmonized Tunisian pronunciation that transcends regional differences, with a focus on consonants, vowels, and assimilation among other patterns. This effort is driven by the need for accurate communication, education, and media in a linguistically diverse nation. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Linguistics

来源：评论

学校读者我要写书评

暂无评论

Performance analysis of complex manufacturing process with sequence data mining technique

International Journal of Control and Automation

引用

International Journal of Control and Automation 2013年第3期6卷 301-312页

作者： Kerdprasop, Kittisak Kerdprasop, Nittaya Data Engineering Research Unit School of Computer Engineering Suranaree University of Technology Thailand

In this paper, we present a sequence analysis method, which is one of the advanced data mining techniques, to identify and extract unique patterns from wafer manufacturing data. Wafer fabrication in the semiconductor industry is one of the most complex manufacturing processes. For such highly complicated operations, maintaining high yields through the statistical process control as a sole monitoring method for quality control is obviously inefficient. We thus investigate the intelligent and semi-automatic technique to help industrial engineers analyzing their production data. Our proposed method has the ability to induce patterns that can reveal and differentiate low performance processes from the normal ones. We also provide program coding of the proposed sequence analysis method, implemented with the R language, for easy experimental repetition.

关键词： Statistical process control

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：