检索结果-内蒙古大学图书馆

Algorithm for Deriving Position Measurement Based on Automatically Saved Image Matching

Transactions of the Korean Institute of Electrical Engineers 2025年第3期74卷 491-499页

作者： Hwang, Dong-Yun Seo, Kyeong-Wook Lee, Min-Ho Kim, Yong-Hun Song, Jin-Woo Dept. of Artificial Intelligence and Robotics Sejong University Korea Republic of Dept. of Intelligent Mechatronics Engineering Sejong University Korea Republic of

In this paper, we propose an algorithm that equips a drone with a monocular camera and a laser range finder to derive position measurements for smooth mission execution and return to the home position, even in situations where GNSS is not available. This algorithm stores images and create an orthoimage when GNSS is available, and from the time the GNSS signal is lost, it matches the current image with the database image based on image feature point matching to determine the drone's current location and to employ return to home mission. We validated the algorithm with data acquired from actual drone flights and confirmed that the proposed algorithm could be used practically and effectively with sufficient accuracy for return to home mission. © The Korean Institute of Electrical Engineers.

关键词： Drones

来源：评论

学校读者我要写书评

暂无评论

Defense and Recovery Strategies for Flash-Based Storage Under Ransomware Attacks: A Survey

Defense and Recovery Strategies for Flash-Based Storage Unde...

引用

2025 International Conference on Electronics, Information, and Communication, ICEIC 2025

作者： Yin, Junhong Kang, Kyungtae Major in. Bio Artificial Intelligence Hanyang University Dept. Computer Science and Engineering Ansan Korea Republic of Hanyang University Dept. Artificial Intelligence Ansan Korea Republic of

ISBN: (纸本)9798331510756

As ransomware evolves, traditional OS-based de-tection mechanisms face growing challenges, particularly from advanced attacks exploiting system vulnerabilities and escalating privileges. The continuous evolution of ransomware attacks has driven researchers to explore storage-level defense strategies, focusing on the potential of Solid-State Drives (SSDs) to enhance data protection and recovery. This paper presents a compre-hensive survey of the current state and future developments in SSD firmware-level ransomware detection and recovery, based on literature published between 2015 and 2024. Firmware-level I/O monitoring helps detect ransomware by identifying the typical 'read-encrypt-overwrite' pattern. This low-level access offers an advantage in detecting ransomware activity. However, conventional detection methods are often prone to false pos-itives. To enhance detection accuracy, some systems integrate machine learning models within the SSD controllers, analyzing I/O request patterns and entropy values of written data to assess potential threats. The Garbage Collection(Gc)mechanism plays a crucial role in data recovery by retaining older versions of data until reclaimed, which can aid in restoring encrypted content. Similarly, backing up legitimate I/O operations incurs performance overhead. To address this, some systems utilize techniques such as neural networks and remote backups to reduce unnecessary performance degradation. Future research should prioritize developing lightweight solutions tailored for SSDs with limited computational resources. Emphasis should be placed on enhancing the timeliness of real-time detection to effectively respond to the rapidly evolving nature of ransomware attacks. © 2025 IEEE.

关键词： Firmware

来源：评论

学校读者我要写书评

暂无评论

Faster R-CNN in Healthcare and Disease Detection: A Comprehensive Review

Faster R-CNN in Healthcare and Disease Detection: A Comprehe...

引用

2025 International Conference on Electronics, Information, and Communication, ICEIC 2025

作者： Tian, Jiawei Lee, Seungho Kang, Kyungtae Major in Bio Artificial Intelligence Hanyang University Dept. of Computer Science and Engineering Ansan-si Korea Republic of Hanyang University Dept. of Applied Artificial Intelligence Ansan-si Korea Republic of Hanyang University Dept. of Artificial Intelligence Ansan-si Korea Republic of

ISBN: (纸本)9798331510756

This review examines the applications, challenges, and prospects of Faster Region-based Convolutional Neural Networks (Faster R-CNN) in healthcare and disease detection. Through a meta-analysis of Web of Science literature from 2017 to 2024, we provide insights into the evolving landscape of Faster R-CNN in medical contexts. The algorithm can be applied to medical image analysis across various modalities, including radiography, computed tomography, magnetic resonance imaging, ultrasound, microscopy, and endoscopy. Its applications extend to optical RGB images for dermatological and surgical uses, as well as broader healthcare areas such as posture detection, medication recognition, and assistive technologies. Despite its success, Faster R-CNN faces challenges in handling subtle abnormalities, addressing class imbalance in medical datasets, ensuring result interpretability, and managing patient privacy. Our analysis reveals dominant application areas, technological advancements, and integration trends with other artificial intelligence technologies. The review highlights Faster R-CNN's significant impact on improving diagnostic accuracy and healthcare delivery while acknowledging the need for continued research to address limitations. Emerging directions include real-time disease detection and advancements in personalized medicine, underscoring Faster R-CNN's potential to further transform healthcare practices and patient outcomes. © 2025 IEEE.

关键词： computerized tomography

来源：评论

学校读者我要写书评

暂无评论

High-Throughput Median Filtering for Large Kernel Sizes on CUDA 17

High-Throughput Median Filtering for Large Kernel Sizes on C...

引用

17th International Conference on Machine Vision, ICMV 2024

作者： Yildirim, Samed Topal, Cihan Dept. Computer Engineering Istanbul Technical University Istanbul Turkey Dept. Artificial Intelligence and Data Engineering Istanbul Technical University Istanbul Turkey

ISBN: (纸本)9781510688278

In this paper, we propose a novel approach to median filtering, a widely employed technique for mitigating specific types of noise in images and signals. While median filtering is effective, its computational demands, especially with larger filter sizes, pose challenges for real-time applications. To address this, we present a highly efficient, optimized, and parallelized median filtering algorithm tailored for CUDA platform. Leveraging a histogram-based method and operating on 8-bit data, our approach outperforms sorting-based alternatives, particularly for larger filter sizes. Quantitative experiments demonstrate substantial performance improvements, with our algorithm achieving speedup up to 20x and 59x compared to the second best GPU-based and CPU-based algorithms for mid and large filter sizes, respectively. This significant enhancement in processing speed makes our algorithm a compelling choice for real-time applications where rapid noise removal is paramount, thereby extending the practical utility of median filtering in various domains. © 2025 SPIE.

关键词： Median filters

来源：评论

学校读者我要写书评

暂无评论

The Utilization of Google Cloud in the Healthcare Industry 2

The Utilization of Google Cloud in the Healthcare Industry

引用

2nd International Conference on Machine Learning and Autonomous Systems, ICMLAS 2025

作者： Kapse, Rishikesh Gourshettiwar, Palash Faculty of Engineering and Technology Dept. of Artificial Intelligence and Data Science Maharashtra Wardha India Faculty of Engineering and Technology Dept. of Computer Science And Medical Engineering Maharashtra Wardha India

ISBN: (纸本)9798331505745

This essay's primary focus is on Google Cloud's use in healthcare and how it affects data management, teamwork, and cost-effectiveness. With a focus on Google Cloud, which improves the processing, storing, and accessibility of medical data, the study examines the trends, advantages, and persistent challenges that come with the increasing usage of cloud computing technology. These consist of remote patient monitoring, 'artificial intelligence,' and 'machine learning' applications in diagnostics, diagnostics, and personalized medicine. Data privacy, security issues, and concerns with the HIPAA incident are among the problems that are discussed along with all the significant elements that could facilitate the cloud migration of healthcare. © 2025 IEEE.

关键词： Theranostics

来源：评论

学校读者我要写书评

暂无评论

Generation of Pseudo X-Ray and Digitally Reconstructed Radiograph from CT images 3

Generation of Pseudo X-Ray and Digitally Reconstructed Radio...

引用

3rd International Conference on Intelligent Data Communication Technologies and Internet of Things, IDCIoT 2025

作者： Arya, R. Kochuvila, Sreeja Dept. of Artificial Intelligence Amrita School of Artificial Intelligence Amrita Vishwa Vidyapeetham Bengaluru India Dept. of Electronics and Communication Engineering Amrita School of Engineering Amrita Vishwa Vidyapeetham Bengaluru India

ISBN: (纸本)9798331527549

Computed Tomography (CT) scanning offers 3D insights into the human body but involves higher radiation exposure than traditional X-Rays, increasing the relevance of 3D reconstructions from 2D data. In this work, datasets of CT- pseudo-X-Ray and CT- Digitally Reconstructed Radiograph (DRR) pairs are generated using ray-sum tracing and projection techniques. This can be helpful for implementing machine learning and deep learning architecture as 2-D data availability is limited. The described approach generates CT–pseudo-X-Ray and CT–Digitally Reconstructed Radiograph (DRR) pairs using ray-sum tracing and projection methods. The pseudo-X-Ray is created by combining bone and soft tissue projections, while the DRR is derived via ray-sum tracing. By retaining the CT images in Digital Imaging and Communications in Medicine (DICOM) format, the created dataset enhances robustness and efficiency, supporting 3D reconstruction—a rapidly advancing field. © 2025 IEEE.

关键词： X ray radiography

来源：评论

学校读者我要写书评

暂无评论

Smart Barcode Scanning using X-ray Technology with computer Vision and Machine Learning 4

Smart Barcode Scanning using X-ray Technology with Computer ...

引用

4th International Conference on Sentiment Analysis and Deep Learning, ICSADL 2025

作者： Vijayaraj, J. Hari Yogith, S. Rithikravivarma, V. Siddharth, S. Easwari Engineering College Dept. of Artificial Intelligence and Data Science Chennai India

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523923

Barcode scanning errors in retail environments contribute to inventory discrepancies, financial losses, and inefficiencies in stock management. Traditional optical barcode scanning methods are prone to failures due to occlusion, poor lighting conditions, and barcode damage. While recent advancements such as computer vision, deep learning-based error detection, and RFID integration have improved scanning accuracy, these approaches still face challenges related to high computational requirements, environmental dependencies, and implementation costs. This paper proposes a novel method utilizing X-ray imaging technology and advanced machine learning algorithms to detect improperly scanned or missing barcodes in real time. The system captures detailed X-ray images of items as they enter retail stores or warehouses and applies deep learning models to analyze barcode visibility. Real-time alerts are generated when scanning anomalies are detected, enabling prompt corrective action. Experimental evaluations demonstrate that the proposed approach significantly improves barcode detection accuracy, enhances inventory management, and reduces operational costs. The findings highlight the potential for widespread deployment in retail and logistics environments, offering a robust solution to barcode scanning challenges. © 2025 IEEE.

关键词： Sales

来源：评论

学校读者我要写书评

暂无评论

LLM driven Natural Language to SQL for Utility Dashboards

LLM driven Natural Language to SQL for Utility Dashboards

引用

2025 IEEE International Students' Conference on Electrical, Electronics and computer Science, SCEECS 2025

作者： Banerji, Jyotirmoy Amin, Abhiraj Rathor, Kapil Mukesh Patel School of Technology Management and Engineering Dept. Artificial Intelligence Mumbai India

ISBN: (纸本)9798331529833

Advancements in Natural Language to Structured Query Language (NL2SQL) systems are improving the accessibility of data stored in relational databases, making it easier for non-technical users to retrieve information using simple language queries. In this paper, we propose an NL2SQL system specifically designed for IoT-driven plant data applications, integrating Large Language Models (LLMs) with a multi-step training and evaluation process. This approach involves developing a context table and accompanying tag datasheet to represent IoT-specific data attributes, fine-tuning a LLM to enhance SQL query generation, and evaluating model performance using the ChatGPT API, comparing generated queries against standard metrics from the Spider dataset. Our methodology aims to address the complexities inherent in NL2SQL tasks involving IoT datasets, where queries are often complex and context specific. This research contributes a scalable and adaptable NL2SQL approach which is suited for IoT data environments. © 2025 IEEE.

关键词： Structured Query Language

来源：评论

学校读者我要写书评

暂无评论

Deepfake Detection of Singing Voices With Whisper Encodings

Deepfake Detection of Singing Voices With Whisper Encodings

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Sharma, Falguni Gupta, Priyanka Centre for Artificial Intelligence Banasthali Vidyapith Jaipur India Dept. of Communication and Computer Engineering The LNM Institute of Information Technology Jaipur India

ISBN: (纸本)9798350368741

The deepfake generation of singing vocals is a concerning issue for artists in the music industry. In this work, we propose a singing voice deepfake detection (SVDD) system, which uses noise-variant encodings of open-AI's Whisper model. As counter-intuitive as it may sound, even though the Whisper model is known to be noise-robust, the encodings are rich in non-speech information, and are noise-variant. This leads us to evaluate Whisper encodings as feature representations for the SVDD task. Therefore, in this work, the SVDD task is performed on vocals and mixtures, and the performance is evaluated in %EER over varying Whisper model sizes and two classifiers-CNN and ResNet34, under different testing conditions. © 2025 IEEE.

关键词： anti-spoofing Deepfake Detection Singing voice deepfake detection Transfer Learning Whisper

来源：评论

学校读者我要写书评

暂无评论

FUN-AD: Fully Unsupervised Learning for Anomaly Detection with Noisy Training Data

FUN-AD: Fully Unsupervised Learning for Anomaly Detection wi...

引用

2025 IEEE/CVF Winter Conference on Applications of computer Vision, WACV 2025

作者： Im, Jiin Son, Yongho Hong, Je Hyeong Hanyang University Dept. Electronic Engineering Korea Republic of Hanyang University Dept. Artificial Intelligence Korea Republic of

ISBN: (纸本)9798331510831

While the mainstream research in anomaly detection has mainly followed the one-class classification, practical industrial environments often incur noisy training data due to annotation errors or lack of labels for new or refurbished products. To address these issues, we propose a novel learning-based approach for fully unsupervised anomaly detection with unlabeled and potentially contaminated training data. Our method is motivated by two observations, that i) the pairwise feature distances between the normal samples are on average likely to be smaller than those between the anomaly samples or heterogeneous samples and ii) pairs of features mutually closest to each other are likely to be homogeneous pairs, which hold if the normal data has smaller variance than the anomaly data. Building on the first observation that nearest-neighbor distances can distinguish between confident normal samples and anomalies, we propose a pseudo-labeling strategy using an iteratively reconstructed memory bank (IRMB). The second observation is utilized as a new loss function to promote class-homogeneity between mutually closest pairs thereby reducing the ill-posedness of the task. Experimental results on two public industrial anomaly benchmarks and semantic anomaly examples validate the effectiveness of FUNAD across different scenarios and anomaly-to-normal ratios. Our code is available at https://***/HY-Vision-Lab/FUNAD. © 2025 IEEE.

关键词： Unsupervised learning

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：