检索结果-内蒙古大学图书馆

Real-Time Road Damage Detection and Infrastructure Evaluation Leveraging Unmanned Aerial Vehicles and Tiny Machine Learning

引用

IEEE Internet of Things Journal 2024年第12期11卷 21347-21358页

作者： Khan, Muhammad Waseem Obaidat, Mohammad S. Mahmood, Khalid Batool, Dania Badar, Hafiz Muhammad Sanaullah Aamir, Muhammad Gao, Wu Comsats University Islamabad Sahiwal Campus Department Of Computer Science Sahiwal57000 Pakistan Superior University Department Of Computer Science Lahore54000 Pakistan The University Of Jordan King Abdullah Ii School Of Information Technology Amman11942 Jordan University Of Science And Technology Beijing School Of Communication And Computing Engineering Beijing100083 China Amity University Amity School Of Engineering And Technology Uttar Pradesh Noida201301 India National Yunlin University Of Science And Technology Graduate School Of Intelligent Data Science Douliu64002 Taiwan University Of Palermo Department Of Engineering Palermo90133 Italy Emerson University Department Of Emerging Computing Technologies Multan60000 Pakistan Northwestern Polytechnical University Center For Bio-Inspired System-on-Chip School Of Computer Science Xi'an710060 China

Road damage detection (RDD) through computer vision and deep learning techniques can ensure the safety of vehicles and humans on the roads. Integrating unmanned aerial vehicles (UAVs) in RDD and infrastructure evaluation (IE) has also emerged as a key enabler, contributing significantly to data acquisition and real-time monitoring of road damages, such as potholes, cracks, and surface anomalies, facilitating proactive maintenance and improved road conditions. These UAVs are low-powered and resource-constrained devices that work autonomously to perform pattern detection and decision making leveraging tiny machine learning (Tiny ML) algorithms. These Tiny ML algorithms are designed to run on edge devices, IoT devices, UAVs, etc. In this study, the RDD2022 data set collected using UAVs and dashboard cameras of vehicles was utilized to train pure and mixed models that exhibit class instance imbalance in certain classes which is addressed by implementing data augmentation as a regularization technique. State-of-the-art two-stage detectors: Faster R-CNN ResNet101 and one-stage detectors: SSD MobileNet V1 FPN, YOLOv5, and Efficientdet D1 are employed. The results indicate that the two-stage detector achieved an impressive mean average precision (mAP) of 88.49% overall and 96.62% for focused classes. Notably, the state-of-the-art Efficientdet D1 approach achieved a competitive mAP of 86.47% overall and 95.12% for focused classes, with significantly lower computational cost. These findings highlight the potential of advanced object detection techniques, particularly Efficientdet D1, to enhance the accuracy and efficiency of RDD systems, thereby improving passenger safety and overall performance. © 2014 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

ZeroMark: towards dataset ownership verification without disclosing watermarks 24

ZeroMark: towards dataset ownership verification without dis...

引用

Proceedings of the 38th International Conference on Neural Information Processing Systems

作者： Junfeng Guo Yiming Li Ruibo Chen Yihan Wu Chenxi Liu Heng Huang Department of Computer Science Institute of Health Computing University of Maryland College Park College of Computing and Data Science Nanyang Technology University

ISBN: (纸本)9798331314385

High-quality public datasets significantly prompt the prosperity of deep neural networks (DNNs). Currently, dataset ownership verification (DOV), which consists of dataset watermarking and ownership verification, is the only feasible solution to protect their copyright by preventing unauthorized use. In this paper, we revisit existing DOV methods and find that they all mainly focused on the first stage by designing different types of dataset watermarks and directly exploiting watermarked samples as the verification samples for ownership verification. As such, their success relies on an underlying assumption that verification is a onetime and privacy-preserving process, which does not necessarily hold in practice. To alleviate this problem, we propose ZeroMark to conduct ownership verification without disclosing dataset-specified watermarks. Our method is inspired by our empirical and theoretical findings of the intrinsic property of DNNs trained on the watermarked dataset. Specifically, ZeroMark first generates the closest boundary version of given benign samples and calculates their boundary gradients under the label-only black-box setting. After that, it examines whether the given suspicious method has been trained on the protected dataset by performing a hypothesis test, based on the cosine similarity measured on the boundary gradients and the watermark pattern. Extensive experiments on benchmark datasets verify the effectiveness of our ZeroMark and its resistance to potential adaptive attacks. The codes for reproducing our main experiments are publicly available at https://***/JunfengGo/***.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Analysis and Performance Evaluation of Credit Card Fraud by Multi-model ML 3

Analysis and Performance Evaluation of Credit Card Fraud by ...

引用

3rd International Conference on Advancement in Electrical and Electronic Engineering, ICAEEE 2024

作者： Islam, Md Aminul Asif Imran, A.T.M. Rahman, Md Habibur Pabel, Md Amran Hossen Mishra, Bhupesh Kumar Basu, Kashinath University of Gloucestershire Data Science School of Engineering and Technology Cheltenham United Kingdom Dhaka Bangladesh University of Chittagong Department of Mathematics Chittagong Bangladesh Wright State University Master of Science in Business Analytics United States United Kingdom Oxford Brookes University Data Science School of Engineering Computing and Mathematics United Kingdom

ISBN: (纸本)9798350388282

Using credit cards is a convenient and efficient payment mechanism. Credit card fraud significantly impacts financial loss, mental health, and the reputation of financial institutions. This study would incorporate an analysis of the preceding statement about eliminating several obstacles associated with the availability of public data, the presence of unbalanced data, the dynamic nature of fraud tendencies, and the prevalence of false alarms. The authors discuss various machine-learning techniques used to detect credit card fraud. Extreme Learning, Decision Trees, Random Forests, Support Vector Machines, Logistic Regression, and XG Boost are among these techniques on the European card benchmark dataset. A comparative evaluation of the effectiveness of machine learning has also been conducted, and precision is increased by incorporating multiple layers. The study's findings indicate significant improvements in several crucial metrics, including accuracy, f1-score, precision, and AUC curves. XGBoost model achieves optimized values of 99.9\% accuracy. The proposed model outperforms contemporary machine learning approaches in credit card fraud detection. It has been observed that the combination of data balancing techniques significantly reduces the occurrence of false negatives. This study has substantial potential for credit card fraud detection applications in the real world, providing an effective and efficient method for addressing this ongoing critical issue. © 2024 IEEE.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Optimizing Diabetic Foot Ulcer Detection: Leveraging SSD and YOLO Architectures 21

Optimizing Diabetic Foot Ulcer Detection: Leveraging SSD and...

引用

21st IEEE International Conference on Smart Communities: Improving Quality of Life using AI, Robotics and IoT, HONET 2024

作者： Azeem, Muhammad Zaman, Muhammad Akhunzada, Adnan Kehkashan, Tanzila Ashraf, Imran Rehman, Abdul The University of Lahore Department of Computer Science Sargodha Pakistan COMSATS University Islamabad Department of Computer Science Islamabad Pakistan University of Doha for Science and Technology Department of Data and Cybersecurity Doha24449 Qatar Universiti Teknologi Malaysia Faculty of Computing Johor Bahru Malaysia The University of Lahore Faculty of Computer Science Sargodha Pakistan

ISBN: (纸本)9798350378078

Diabetic foot ulcer (DFU) are a severe complication in diabetic patients, requiring accurate and early diagnosis to prevent further complications such as infection or amputation. To address this, Research employed YOLO and SSD models for automated DFU detection from medical images, comparing their performance based on key evaluation metrics, including accuracy, precision, recall, and F1-score. Results demonstrate that both models perform exceptionally well, with SSD achieving an accuracy of 98.7%, surpassing YOLO's 97.5%. Additionally, SSD exhibited superior precision (99%), recall (98.9%), and F1-score (99.1%), indicating its robustness in detecting ulcerated regions of varying sizes and stages. These results were further contextualized through a comparison with previous studies, where SSD outperformed several CNN-based models, which achieved accuracy ranging from 93.7% to 97.0%. The SSD model's multi-scale object detection capabilities, combined with transfer learning and data augmentation techniques, contributed to its high performance. Overall, this study demonstrates the potential of deploying SSD and YOLO models in clinical settings for early and accurate DFU detection, paving the way for more reliable and timely interventions in diabetic care. The findings underline the significance of advanced object detection methods, offering promising avenues for improving patient outcomes in medical imaging © 2024 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Predicting Cervical Cancer Using Deep Learning Mutation-based Atom Search Optimization (MASO) Algorithm 2

Predicting Cervical Cancer Using Deep Learning Mutation-base...

引用

2nd IEEE International Conference on Trends in Quantum computing and Emerging Business Technologies, TQCEBT 2024

作者： Mayuri, A.V.R. Selvekumar Subash, K. Aeron, Anurag Vit Bhopal University School of Computing Science and Engineering Madhya Pradesh Sehore India Bahir Dar University Bahir Dar Institute of Technology Department of Electrical and Computer Engineering Ethiopia Department of Data Science Trichy India Miet Meerut Department of Computer Science and Engineering Uttar Pradesh India

ISBN: (纸本)9798350384277

Worldwide, women are compressed by cervical cancer, which is a prevalent malignancy. This disease, which is currently the fourth leading cause of death for women, shows no symptoms when it first arises. Cells that cause cervical cancer multiply gradually at the cervix. If this cancer is detected early enough, treatment can be effective. Presently, it is difficult for medical workers to detect this kind of cancer before it spreads rapidly. This study used a number of machine learning classification algorithms with risk markers to predict cervical cancer. These study suggestions mutation-based Atom Search Optimization (MASO) with a Deep Convolutional Neural Network (DCNN) to offer a original method for automated diagnosis of uterine cervical cancer. Though DCNN is used to extract features from medical imaging data, MASO is used to increase the optimization of the diagnostic model. Composed, these technologies have the potential to rise cervical cancer diagnosis accuracy and efficiency, offering an automated approach for early detection and prompt intervention. Our proposed model established a strong 95% accuracy rate for generalization. This study explores the difficulties caused by missing values and class imbalance in the specific dataset with the goal of assisting medical professionals in the early detection of cervical cancer and improving treatment for people that are impacted by the illness. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Fuzzy-Based Optimization Techniques for Segmenting the Tumors in Multimodal MRI Images

引用

Operations Research Forum 2025年第1期6卷 1-27页

作者： Alagarsamy, Saravanan Nagarajan, D. Govindaraj, Vishnuvardhan Department of Information Technology Sri Sivasubramaniya Nadar College of Engineering Rajiv Gandhi Salai (OMR) Tamilnadu Chennai India Department of Mathematics Rajalakshmi Institute of Technology Chennai India Division of Data Science of School of Computing Science and Education VIT Bhopal University Kothrikalan Madhya Pradesh Sehore 466114 India

Accurate brain tumor prediction is essential when approaching the field of health care;wherever accuracy in decision-making is crucial, the issues also need to be resolved right away. Many artificial intelligence and machine learning-based techniques have been developed in the last couple of years in the field of healthcare. The intention of this work is to create an algorithm that combines the features of Crow Search (CS) and Interval Type-II Fuzzy Logic System (IT2FLS) algorithms to distinguish the area of tumor from complex brain tissues. The ability of oncologists to make decisions is critical for any therapy sequence to be successful, and the methodology proposed in the research work considerably influences a conclusion through technology interruption. The suggested approach is flexible enough to work with a diversity of image sequences found in the BRATS Challenge 2020 dataset that presents varying degrees of obstacles, challenges, and difficulties in locating the anomalous regions, and it produces improved demarcation results that have been instinctively evaluated and supported by Dice score (98 ± 1.3), specificity (98 ± 1.4), and sensitivity (98 ± 1.7) as average metrics. The motive of this study is the enhancement of the visual perception of oncologists, which gives them improved insight into and comprehension of the condition of the patient. © The Author(s), under exclusive licence to Springer Nature Switzerland AG 2025.

关键词： Crow Search (CS) Interval Type-II Fuzzy Logic System (IT2FLS) Magnetic Resonance Imaging (MRI) Tumor segmentation

来源：评论

学校读者我要写书评

暂无评论

Securing 3D Point and Mesh Fog data Using Novel Chaotic Cat Map

引用

Computers, Materials & Continua 2023年第3期74卷 6703-6717页

作者： K.Priyadarsini Arun Kumar Sivaraman Abdul Quadir Md Areej Malibari Department of Data Science and Business Systems School of ComputingCollege of Engineering and TechnologySRM Institute of Science and TechnologyKattankulathurChennai603203India School of Computer Science and Engineering Vellore Institute of TechnologyChennai600127India Department of Industrial and Systems Engineering College of EngineeringPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia

With the rapid evolution of Internet technology,fog computing has taken a major role in managing large amounts of *** major concerns in this domain are security and ***,attaining a reliable level of confidentiality in the fog computing environment is a pivotal *** different types of data stored in the fog,the 3D point and mesh fog data are increasingly popular in recent days,due to the growth of 3D modelling and 3D printing ***,in this research,we propose a novel scheme for preserving the privacy of 3D point and mesh fog *** Cat mapbased data encryption is a recently trending research area due to its unique properties like pseudo-randomness,deterministic nature,sensitivity to initial conditions,ergodicity,*** boost encryption efficiency significantly,in this work,we propose a novel Chaotic Cat *** sequence generated by this map is used to transform the coordinates of the fog *** improved range of the proposed map is depicted using bifurcation *** quality of the proposed Chaotic Cat map is also analyzed using metrics like Lyapunov exponent and approximate *** also demonstrate the performance of the proposed encryption framework using attacks like brute-force attack and statistical *** experimental results clearly depict that the proposed framework produces the best results compared to the previous works in the literature.

关键词： Chaotic cat map fog computing encryption 3D point fog 3D mesh

来源：评论

学校读者我要写书评

暂无评论

Towards Techniques for Updating Virtual Knowledge Graphs 17

Towards Techniques for Updating Virtual Knowledge Graphs

引用

17th International Rule Challenge and 7th Doctoral Consortium @ RuleM+RR, RuleML+RR-Companion 2023

作者： Wandji, Romuald Esdras Šimkus, Mantas Calvanese, Diego Department of Computing Science Umeå Universitet Umeå Sweden Research Centre for Knowledge and Data Faculty of Engineering Free University of Bozen-Bolzano Bolzano Italy

The field of Virtual Knowledge Graphs (VKGs) continues to grow in both academic and applied contexts. Yet, the issue of updates in VKG systems has not yet received adequate attention, although it is crucial to manage data modifications at the data source level through the lens of an ontology. In this paper, we focus on VKGs whose ontology is specified in the lightweight ontology language DL-LiteA, and we propose diverse settings and research directions we intend to explore to address the challenge of translating ontology-based updates into updates at the level of data sources. We also pay attention to the important problem of automated analysis of mappings, which plays a major role when it comes to reformulating ontology-based update requests into update requests over the data sources. © 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Ontology

来源：评论

学校读者我要写书评

暂无评论

Deep Neural Network Approach on Phishing Websites Detection, Identification and Proactive Prevention 8

Deep Neural Network Approach on Phishing Websites Detection,...

引用

8th IEEE International Conference on Computational System and Information Technology for Sustainable Solutions, CSITSS 2024

作者： Soundarya, K. Siddaraju, H.S. Tekchandani, Hitesh Bhardwaj, Navodit Rani, Pooja Joshi, Pallavi Amrita School of Computing Amrita Vishwa Vidyapeetham Department of Computer Science Mysuru India Koneru Lakshmaiah Education Foundation Artificial Intelligence and Data Science Hyderabad India Researcher at Food and Drug Administration MD United States

ISBN: (纸本)9798331505462

Phishing is a prevalent type of cyberattack that involves posing as a trustworthy source in an email message to fraudulently attempt to get personal information such as usernames, passwords, and bank account information. Strong detection and prevention methods are required to address the phishing issue. The objective is to develop efficient algorithms that use machine learning (ML) and data mining methods to accurately detect and categorize phishing URLs. This initiative aims to support ongoing initiatives to strengthen cybersecurity and shield consumers from the risks associated with phishing scams. It can be observed from the results that the Deep neural network model gives 95% of accuracy and 96% of precision for classification and detection of phishing websites. © 2024 IEEE.

关键词： Phishing

来源：评论

学校读者我要写书评

暂无评论

Semantic Querying of Integrated Raster and Relational data: A Virtual Knowledge Graph Approach 17

Semantic Querying of Integrated Raster and Relational Data: ...

引用

17th International Rule Challenge and 7th Doctoral Consortium @ RuleM+RR, RuleML+RR-Companion 2023

作者： Ghosh, Arka Šimkus, Mantas Calvanese, Diego Department of Computing Science Umeå Universitet Umeå Sweden Research Centre for Knowledge and Data Faculty of Engineering Free University of Bozen-Bolzano Bolzano Italy

Ontology-based data access (OBDA) facilitates access to heterogeneous data sources through the mediation of an ontology (e.g. OWL), which captures the domain of interest and is connected to data sources through a declarative mapping. In our study, large, heterogeneous earth observational (EO) data, known as raster data, and geometrical data, known as vector data, are considered as (heterogeneous) data sources. Raster data represent, e.g., Earth's natural phenomena, such as surface temperature, elevation, or air pollution, as multidimensional arrays. In contrast, vector data depict, e.g., locations, networks, or regions on Earth, using geometries. Domain experts, such as earth scientists and GIS practitioners, still struggle to undertake advanced studies by querying large raster and vector data in an integrated way because, unlike relational data, they come in diverse formats and different data structures. In our approach to integration, we use a geospatial extension of an RDBMS to represent vector data as relational data, and a domain-agnostic array DBMS to handle raster data. Our aim is to extend the OBDA paradigm to effectively deal with relational, vector, and raster data in a combined way, while leveraging the built-in capabilities of data management tools relevant to each type of data. We also plan to develop techniques to calculate on the fly for each user query posed over the ontology an optimal query plan that exploits, at best, the query processing capabilities of each tool, while limiting costly data transfer operations between tools. © 2023 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).

关键词： Semantics

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：