检索结果-内蒙古大学图书馆

Image-Based Automatic Energy Meter Reading Using Deep Learning

computers, Materials & Continua 2023年第1期74卷 203-216页

作者： Muhammad Imran Hafeez Anwar Muhammad Tufail Abdullah Khan Murad Khan Dzati Athiar Ramli Institute of Computer Sciences and Information Technology Faculty of Management and Computer SciencesThe University of AgriculturePeshawarPakistan Department of Electrical and Computer Engineering COMSATS University IslamabadAttock CampusAttock CityPakistan Kuwait College of Science and Technology Kuwait CityKuwait School of Electrical&Electronic Engineering USM Engineering CampusUniversiti Sains MalaysiaPenang14300Malaysia

We propose to perform an image-based framework for electrical energy meter *** aim is to extract the image region that depicts the digits and then recognize them to record the consumed *** the readings of serial numbers and energy meter units,an automatic billing system using the Internet of Things and a graphical user interface is deployable in a real-time ***,such region extraction and character recognition become challenging due to image variations caused by several factors such as partial occlusion due to dust on the meter display,orientation and scale variations caused by camera positioning,and non-uniform illumination caused by *** this end,our work evaluates and compares the stateof-the art deep learning algorithm You Only Look Once(YOLO)along with traditional handcrafted features for text extraction and *** image dataset contains 10,000 images of electrical energymeters and is further expanded by data augmentation such as in-plane rotation and scaling tomake the deep learning algorithms robust to these image *** training and evaluation,the image dataset is annotated to produce the ground truth of all the ***,YOLO achieves superior performance over the traditional handcrafted features with an average recognition rate of 98%for all the *** proves to be robust against the mentioned image variations compared with the traditional handcrafted *** proposed method can be highly instrumental in reducing the time and effort involved in the currentmeter reading,where workers visit door to door,take images ofmeters and manually extract readings from these images.

关键词： Convolutional neural network object localization machine learning

来源：评论

学校读者我要写书评

暂无评论

ECDSA-based tamper detection in medical data using a watermarking technique

引用

International Journal of Cognitive Computing in engineering 2024年第1期5卷 78-87页

作者： Ch, Rupa Vivek K, Naga Srivastava, Gautam Reddy Gadekallu Department CSE VR Siddhartha Engineering College Andhra Pradesh Vijayawada520 007 India Dept of Math and Computer Science Brandon University MB Canada Dept. of Computer Science and Math Lebanese American University Beirut1102 Lebanon Centre of Research Impact and Outcome Chitkara University Institute of Engineering and Technology Chitkara Punjab Rajpura140401 India Lebanese American University Department of Electrical and Computer Engineering Byblos Lebanon Lovely Professional University Division of Research and development Phagwara India Jiaxing University College of Information Science and Engineering Jiaxing China

Telemedicine is a form of healthcare delivery that employs communication technology to provide medical care to patients remotely. The use of telemedicine has seen a significant increase in recent years, presenting challenges such as patient privacy, data security, the need for reliable communication technology, and the potential for misdiagnosis without a physical examination. Digital Watermarking can assist in addressing such issues by incorporating a unique identifier into an image that can be used to authenticate its validity. To tackle these issues, this study proposes a robust digital watermarking approach tailored to brain medical images, combining hashing, the Elliptic Curve Digital Signature Algorithm (ECDSA), and the Integer Wavelet Transform-Discrete Cosine Transform (IWT-DCT). This method utilizes the Secure Hash Algorithm (SHA-256) to first segment the brain's Region of Interest (RoI). Subsequently, the hashed RoI, along with an ECDSA signature, is embedded into the high-frequency sub-bands of the medical image using IWT-DCT. The embedding process strategically alters the coefficients of the high-frequency sub-bands to accommodate the signature while minimizing perceptual distortion. The technique leverages the robustness of transformed-domain image watermarking techniques against various attacks and combines it with SHA-256 for integrity and ECDSA for authentication purposes. The results demonstrate that the suggested approach is robust to a variety of image processing techniques, including noise addition, filtering, and compression while maintaining high levels of imperceptibility. Key metrics such as the Peak Signal-to-Noise Ratio (PSNR), Mean Squared Error (MSE), and Structural Similarity Index (SSIM) were used to evaluate performance. The suggested strategy exhibited a substantial improvement over existing methods. The PSNR increased to 68.67, indicating higher image quality, while the MSE reduced to 0.96, demonstrating closer pixel values to the or

关键词： Mean square error

来源：评论

学校读者我要写书评

暂无评论

The Effectiveness of CNN-Based Models in Macroscopic Motion Direction Detection 6

The Effectiveness of CNN-Based Models in Macroscopic Motion ...

引用

6th International Conference on Robotics and computer Vision, ICRCV 2024

作者： Qiu, Zhiyu Chen, Tianqi Hua, Yuxiao Todo, Yuki Graduate School of Natural Science & Technology Kanazawa University Division of Electrical Engineering and Computer Science Kanazawa Japan Kanazawa University Faculty of Electrical Information and Communication Engineering Kanazawa Japan

ISBN: (纸本)9798331527426

Convolutional Neural Networks (CNNs) have revolutionized the field of computer vision, offering significant advancements in image and video recognition, classification, and segmentation tasks. Leveraging hierarchical pattern recognition capabilities, CNNs have demonstrated superior performance over traditional machine learning methods. One particularly critical application of CNNs is in motion direction detection, which is essential for enhancing security in surveillance systems, ensuring safe navigation in autonomous vehicles, and providing valuable insights in sports analytics. In this study, our primary goal was to investigate how different types of input data impact the accuracy of motion direction detection. We applied five distinct preprocessing methods to test the effect of input data on accuracy. Furthermore, we conducted experiments in various object and background scenarios to evaluate adaptability of simple 4-layer CNN, EfficientNetBO and ResNet50 in detecting motion direction. Finally, we tested the noise resistance of these three models. The results indicate that CNNs are capable of extracting the macroscopic motion direction of objects in images. Our study highlights the potential of CNN-based models for efficient and effective motion direction detection. © 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

Implementation and experimental evaluation of dynamic capabilities in wireless body area networks: different setting parameters and environments

引用

Journal of engineering and Applied science 2023年第1期70卷 1-25页

作者： Thippun, Pitchakron Sasiwat, Yoschanin Buranapanichkit, Dujdow Booranawong, Apidet Jindapetch, Nattha Saito, Hiroshi Department of Electrical Engineering Faculty of Engineering Prince of Songkla University Hat Yai90110 Thailand Division of Computer Engineering The University of Aizu Aizu-Wakamatsu965-8580 Japan

It is well known that a wireless body area network (WBAN) is a special proposed wireless sensor network (WSN) that can assist in monitoring physiological signals for the evaluation and planning of patient treatment. One of the most challenging issues for WBANs is communication reliability, with acceptable communication efficiency and packet loss. To obtain such network characteristics, collision-free data transmission in networks of wireless sensor nodes is an interesting research problem. In this paper, the experiments of dynamic capabilities in several WBAN scenarios are focused, where the novelty and major contribution of our tests is that the effects of packet inter-arrival times, packet sizes, and the number of nodes deployed in the network, including human movements, indoor and outdoor environments, and transmitter and receiver positions, are all taken into consideration and evaluated. This is achieved by implementing the WBAN using IEEE 802.15.4 low-power sensor nodes. Experimental results illustrate the significant factors that impact the communication reliability of WBANs as measured by the packet delivery ratio (PDR). The experimental results show that the diverse environment testbed can affect network performance for WBAN data transmission. Our findings also show that the best network reliability needs to be set at more than 15 ms in packet interval time to achieve over 90% PDR for every test scenario. More details of the experimental results related to WBAN reliability obtained from all test cases are also discussed and summarized in the paper. To the best of our knowledge, our findings can be useful for users and researchers to consider the optimal point for WBAN setting and configuration to achieve the communication reliability requirements and also to deploy and develop a more reliable WBAN system. © 2023, The Author(s).

关键词： Data transfer

来源：评论

学校读者我要写书评

暂无评论

A FSS Inverse Design Method Based on Output Space Mapping Enhanced ECM

A FSS Inverse Design Method Based on Output Space Mapping En...

引用

2024 International Applied Computational Electromagnetics Society Symposium, ACES-China 2024

作者： Tian, Ziheng Zhang, Juntao Kuang, Yu Cheng, Qingsha S. Southern University of Science and Technology Department of Electrical and Electronic Engineering Shenzhen China National University of Singapore Department of Electrical and Computer Engineering 117583 Singapore

ISBN: (纸本)9798350355581

This paper proposes an inverse design method for frequency selective surfaces (FSS) based on an equivalent circuit model (ECM) and output space mapping (OSM) technique. The method establishes an OSM enhanced ECM model (forward model) to replace expensive electromagnetic (EM) simulations for FSS shielding effectiveness (SE) evaluation. Furthermore, an inverse model is proposed to predict FSS unit cell structure parameter values using specification-driven feature point sampling method. Experiment results demonstrate that FSS designs with shielding frequency between 2-18GHz can be efficiently predicted and verified by EM simulation. © 2024 IEEE.

关键词： Frequency selective surfaces

来源：评论

学校读者我要写书评

暂无评论

Optimizing Different Methods of Fabricating 3D Printed UV Blocking Sheets for Solar Cells 1

Optimizing Different Methods of Fabricating 3D Printed UV Bl...

引用

1st Renewable Energy Technologies and Modern Communications Systems: Future and Challenges Conference, RETMCS 2024

作者： Alhindi, Amjad Fawzi Elatab, Nazek Baawad, Lujain Khalid Bahabry, Rabab R. Allam, Fahad Laboratory Electrical and Computer Engineering Computer Electrical Mathematical Science and Engineering Thuwal Saudi Arabia College of Science Department of Physical Sciences Jeddah21589 Saudi Arabia Materials Science and Engineering Department King Fahd University of Petroleum and Minerals Dhahran31261 Saudi Arabia

ISBN: (纸本)9798331519445

This study investigates the fabrication of innovative UV-blocking sheets that effectively transmit visible light while simultaneously obstructing harmful ultraviolet (UV) radiation, utilizing Cerium Oxide (CeO2) and Zinc Oxide (ZnO) nanoparticles as UV-blocking agents. Three distinct fabrication methods were explored: The Powder Method, where nanoparticles were incorporated directly into polydimethylsiloxane (PDMS);the Ball milling Mixture method, which involved mixing nanoparticles through ball milling;and the Solvent-Based Mixture method, where nanoparticles were dissolved, mixed, and dried;before incorporation into PDMS. Samples were produced using a 3D printing technique, which is Direct Ink Writing (DIW) and characterized for their optical properties via UV-Vis spectroscopy. The results indicated CeO2 demonstrated superior UV-blocking capabilities (a maximum blockage of ~99%) compared to ZnO, while maintaining high transmittance for visible light (a maximum of ~90%). Notably, the solvent-based method exhibited better results than the ball milling method;however, the powder method had the best results. Therefore, this investigation offers a promising solution for solar panel protection that aligns with Saudi Arabia's Vision 2030 energy objectives. The relentless exposure of devices, particularly solar cells, to harmful UV radiation poses a formidable challenge and leads to annual degradation rates of between −0.8% and −4.9%, besides other factors. These findings underscore the critical importance of developing effective UV-blocking solutions. ©2024 IEEE.

关键词： Polydimethylsiloxane

来源：评论

学校读者我要写书评

暂无评论

A Saliency Based Image Fusion Framework for Skin Lesion Segmentation and Classification

引用

computers, Materials & Continua 2022年第2期70卷 3235-3250页

作者： Javaria Tahir Syed Rameez Naqvi Khursheed Aurangzeb Musaed Alhussein Department of Computer Science COMSATS University IslamabadIslamabad CampusPakistan Department of Electrical and Computer Engineering COMSATS University IslamabadWah CampusPakistan Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityRiyadh11543Saudi Arabia

Melanoma,due to its higher mortality rate,is considered as one of the most pernicious types of skin cancers,mostly affecting the white *** has been reported a number of times and is now widely accepted,that early detection of melanoma increases the chances of the subject’s ***-aided diagnostic systems help the experts in diagnosing the skin lesion at earlier stages using machine learning *** thiswork,we propose a framework that accurately segments,and later classifies,the lesion using improved image segmentation and fusion *** proposed technique takes an image and passes it through two methods simultaneously;one is the weighted visual saliency-based method,and the second is improved HDCT based saliency *** resultant image maps are later fused using the proposed image fusion technique to generate a localized lesion *** resultant binary image is later mapped back to the RGB image and fed into the Inception-ResNet-V2 pre-trained model-trained by applying transfer *** simulation results show improved performance compared to several existing methods.

关键词： Skin lesion segmentation image fusion saliency detection skin lesion classification deep neural networks transfer learning

来源：评论

学校读者我要写书评

暂无评论

Multi-Server Stable Rendezvous for the Metaverse

Multi-Server Stable Rendezvous for the Metaverse

引用

2023 IEEE International Conference on Metaverse Computing, Networking and Applications, MetaCom 2023

作者： Su, Ningxin Li, Baochun Li, Bo University of Toronto The Department of Electrical and Computer Engineering Canada Hong Kong University of Science and Technology The Department of Computer Science Hong Kong

ISBN: (纸本)9798350333336

When the real world and the digital world meet, they create a shared virtual space called the metaverse, involving multiple virtual communities in which users interact with one another in a highly immersive and interactive fashion. Due to the pressing need for scalability when handling millions of users in the metaverse regardless of the applications and services provided, it is imperative to deploy multiple servers across geographically distributed datacenters around the world. In this paper, we envision, design, and implement a new rendezvous service between a large number of users and a collection of geographically distributed servers, taking into account the latencies and pair-wise network bandwidth between the users and the servers, as well as bandwidth and processing capacity constraints on the servers themselves when handling the users. Our new rendezvous service is designed with simplicity and efficiency as its primary objective, and uses a revised design of the Deferred Acceptance algorithm to guarantee a stable matching between the users in the metaverse and its servers. As a case study, our rendezvous service has been implemented in the context of a federated learning application. © 2023 IEEE.

关键词： Bandwidth

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Certain Convolutional Neural Network Architectures for X-ray Image Analysis in Bone Fracture Detection and Identification

A Comparative Study of Certain Convolutional Neural Network ...

引用

2024 International Conference on Artificial Intelligence, computer, Data sciences, and Applications, ACDSA 2024

作者： Kabir, Mashrur Tahiti, Tasnia Jasim Ahsan Prome, Tasnim Brac University Department of Electrical and Electronic Engineering Dhaka Bangladesh Brac University Department of Computer Science and Engineering Dhaka Bangladesh

ISBN: (纸本)9798350394528

This research systematically evaluates the performance of diverse Convolutional Neural Network (CNN) architectures in enhancing the accuracy of bone fracture detection in medical imaging. The study aims to understand the intricate nuances exhibited by CNNs when analyzing X-ray images, high-lighting the significance of a robust framework with high sensitivity and specificity. To address the issue of imbalanced datasets, a carefully preprocessed, normalized, and augmented dataset from multiple medical institutions is utilized. The implementation of CNN architectures, such as ResNet, VGGNet, and InceptionNet, involves meticulous configuration and hyperparameter tuning to optimize feature extraction in this complex problem domain. Through extensive experimentation and thorough examination of metrics including accuracy, F1 score, sensitivity, and specificity, the efficacy of each architecture in identifying fractures and distinguishing them from benign anomalies in X-ray images is uncovered. The analysis provides a comprehensive understanding of the strengths and limitations of each architecture, enabling well-informed decisions regarding their suitability in clinical settings. This research represents a significant advancement in the field of bone fracture detection in medical imaging, offering valuable insights into the transformative potential of CNN architectures in improving diagnostic accuracy and informing clinical decision-making. © 2024 IEEE.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Pruning then Reweighting: Towards Data-Efficient Training of Diffusion Models

Pruning then Reweighting: Towards Data-Efficient Training of...

引用

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

作者： Li, Yize Zhang, Yihua Liu, Sijia Lin, Xue Department of Electrical and Computer Engineering Northeastern University Boston United States Department of Computer Science and Engineering Michigan State University East Lansing United States

ISBN: (纸本)9798350368741

Despite the remarkable generation capabilities of Diffusion Models (DMs), conducting training and inference remains computationally expensive. Previous works have been devoted to accelerating diffusion sampling, but achieving data-efficient diffusion training has often been overlooked. In this work, we investigate efficient diffusion training from the perspective of dataset pruning. Inspired by the principles of data-efficient training for generative models such as generative adversarial networks (GANs), we first extend the data selection scheme used in GANs to DM training, where data features are encoded by a surrogate model, and a score criterion is then applied to select the coreset. To further improve the generation performance, we employ a class-wise reweighting approach, which derives class weights through distributionally robust optimization (DRO) over a pre-trained reference DM. For a pixel-wise DM (DDPM) on CIFAR-10, experiments demonstrate the superiority of our methodology over existing approaches and its effectiveness in image synthesis comparable to that of the original full-data model while achieving the speed-up between 2.34× and 8.32×. Additionally, our method could be generalized to latent DMs (LDMs), e.g., Masked Diffusion Transformer (MDT) and Stable Diffusion (SD), and achieves competitive generation capability on ImageNet. © 2025 IEEE.

关键词： data reweighting data-efficient training diffusion model

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：