检索结果-内蒙古大学图书馆

deep Guided Attention Network for Joint Denoising and Demosaicing in real image

Chinese Journal of Electronics 2024年第1期33卷 303-312页

作者： Tao ZHANG Ying FU Jun ZHANG School of Computer Science and Technology Beijing Institute of Technology Advanced Reasearch Institute of Multidisciplinary Science Beijing Institute of Technology

Denoising(DN) and demosaicing(DM) are the first crucial stages in the image signal processing pipeline. Recently, researches pay more attention to solve DN and DM in a joint manner, which is an extremely undetermined inverse problem. Existing deep learning methods learn the desired prior on synthetic dataset, which limits the generalization of learned network to the real world data. Moreover, existing methods mainly focus on the raw data property of high green information sampling rate for DM, but occasionally exploit the high intensity and signalto-noise(SNR) of green channel. In this work, a deep guided attention network(DGAN) is presented for real image joint DN and DM(JDD), which considers both high SNR and high sampling rate of green information for DN and DM, respectively. To ease the training and fully exploit the data property of green channel, we first train DN and DM sub-networks sequentially and then learn them jointly, which can alleviate the error accumulation. Besides, in order to support the real image JDD, we collect paired raw clean RGB and noisy mosaic images to conduct a realistic dataset. The experimental results on real JDD dataset show the presented approach performs better than the state-of-the-art methods, in terms of both quantitative metrics and qualitative visualization.

关键词： image denoising image demosaicing Joint processing Guided attention Paired real dataset

来源：评论

学校读者我要写书评

暂无评论

deep learning-Based Optimization of Cloud Enterprise Resource Planning (ERP) Systems for Adaptive Decision Support and Management Effectiveness Analysis

引用

IEEE ACCESS 2024年 12卷 193402-193415页

作者： Zhang, Li-Sen George Washington Univ Sch Business Washington DC 20052 USA

Modern enterprise resource planning (ERP) systems face the challenge of handling massive amounts of data and supporting real-time decision-making. With the rapid changes in the market environment, traditional ERP systems are limited in their ability to make adaptive decisions. This study aims to address this issue by integrating deep learning techniques to enhance the management effectiveness of ERP systems. The study uses RNNs, CNNs and DRL models for time series prediction, image recognition and resource optimisation, respectively. The experimental results show that RNN achieves 95% accuracy in demand forecasting, CNN 98% accuracy in image recognition, and DRL achieves more than 10% cost savings in resource optimisation. The integrated ERP system achieved a 42.86% reduction in order processing time, a 25% improvement in inventory turnover, an 8% reduction in operating costs, and a 15% improvement in employee satisfaction. This study demonstrates the effectiveness of deep learning to enhance decision support in ERP systems and provides suggestions for future directions of improvement.

关键词： Optimization Accuracy Convolutional neural networks Adaptive systems Decision making Business Recurrent neural networks Predictive models Enterprise resource planning Adaptation models Thesaurus deep learning cloud-based enterprise resource planning system adaptive decision support management effectiveness

来源：评论

学校读者我要写书评

暂无评论

Syflo: augmenting yolo for real-time health monitoring of electric assets in power transmission lines

引用

JOURNAL OF real-time image processing 2025年第1期22卷 1-15页

作者： Sankuri, Raja Sekhar Sristy, Nagesh Bhattu Karri, Sri Phani Krishna Natl Inst Technol Andhra Pradesh Dept Elect Engn Tadepalligudem India Natl Inst Technol Andhra Pradesh Dept Comp Sci & Engn Tadepalligudem India

Sustainable transmission of electrical energy to consumers across regions relies heavily on the integrity of power transmission lines and continuous monitoring of assets is crucial for maintaining system reliability. Unmanned aerial vehicles have revolutionized defect identification in real-time and accessibility, even in difficult-to-reach geographical landscapes, thereby improving image-based inspections. This work introduces semisupervised Yolo with focal loss function (SYFLo), a novel method that augments YOLO for real-time health monitoring of electric assets in power transmission lines. SYFLo integrates the focal loss function with semi-supervised learning to effectively address the lack of abundant labeled data, data imbalances and enhance detection accuracy. Additionally, it improves data generalizability across a wide range of images, ensuring robust performance despite varied image backgrounds. By leveraging YOLOv8, SYFLo significantly improves fault identification, achieving a detection accuracy of 96.5% and an FPS of 16.39. Experimental results demonstrate the impact of the proposed approach, highlighting its potential to enhance the reliability of power transmission line monitoring. These findings underscore the importance of integrating advanced deep learning techniques with innovative loss functions to address common challenges in real-time health monitoring systems.

关键词： deep learning Power grid inspection Defect detection Semi-supervised learning Focal loss function

来源：评论

学校读者我要写书评

暂无评论

Online Recognition of Manually Spray-Printed Ladle Numbers in Complex Dynamic Environments

引用

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT 2024年 73卷 1页

作者： Meng, Hongji Ma, Hancong Yang, Jian Hu, Zhenwei Fan, Jia He, Qing Northeastern Univ Coll Informat Sci & Engn Dept Intelligent Detect Shenyang 110819 Peoples R China HBIS Grp HanSteel Co Handan Peoples R China

During the converter process, it is crucial to automatically identify and record ladle numbers to track steel product quality and enhance automation levels. However, the steelmaking environment presents several challenges, including intricate ladle scheduling, varying lighting conditions, severe background interference, and significant disparities between manually spray-printed ladle number characteristics and publicly available datasets. The combination of these problems makes it challenging to perform accurate and real-time ladle number identification. In response, this article suggests an automatic ladle number recognition approach based on deep learning and image processing. First, a double-region object detection model based on YOLOv5 is employed to capture keyframe images of the ladle to be identified from the video stream. Then, a method that can enable the acquisition of an accurate region of ladle numbers in sophisticated industrial settings is proposed to address the distortion of numerical features caused by lighting variations and background interference in industrial environments. Last, leveraging the proprietary dataset found and a ladle number recognition model integrating CNN and multiframe image fusion is designed, developing multithreading design and image queue management to ensure real-time and accurate ladle number recognition. In this study, the video data of a steel plant is used for testing. Through testing 176 steelmaking production cycles, all ladle numbers are accurately identified prior to finishing charging molten iron, indicating the high accuracy and real-time capability of the recognition system.

关键词： image recognition Feature extraction YOLO real-time systems Streaming media Iron image segmentation Complex dynamic environments deep learning image processing ladle number recognition online recognition

来源：评论

学校读者我要写书评

暂无评论

Demystifying the Hypercomplex: Inductive biases in hypercomplex deep learning

引用

IEEE SIGNAL processing MAGAZINE 2024年第3期41卷 59-71页

作者： Comminiello, Danilo Grassucci, Eleonora Mandic, Danilo P. Uncini, Aurelio Sapienza Univ Rome Dept Informat Engn Elect & Telecommun I-00184 Rome Italy Imperial Coll London Dept Elect & Elect Engn London SW7 2BT England

Hypercomplex algebras have recently been gaining prominence in the field of deep learning owing to the advantages of their division algebras over real vector spaces and their superior results when dealing with multidimensional signals in real-world 3D and 4D paradigms. This article provides a foundational framework that serves as a road map for understanding why hypercomplex deep learning methods are so successful and how their potential can be exploited. Such a theoretical framework is described in terms of inductive bias, i.e., a collection of assumptions, properties, and constraints that are built into training algorithms to guide their learning process toward more efficient and accurate solutions. We show that it is possible to derive specific inductive biases in the hypercomplex domains, which extend complex numbers to encompass diverse numbers and data structures. These biases prove effective in managing the distinctive properties of these domains as well as the complex structures of multidimensional and multimodal signals. This novel perspective for hypercomplex deep learning promises to both demystify this class of methods and clarify their potential, under a unifying framework, and in this way, promotes hypercomplex models as viable alternatives to traditional real-valued deep learning for multidimensional signal processing.

关键词： deep learning Training data Multidimensional signal processing Three-dimensional displays Algebra image processing Hypercomplex

来源：评论

学校读者我要写书评

暂无评论

Experimental Analysis of Malarial Parasites Infection by using deep learning based on Medical image processing Logic 3

Experimental Analysis of Malarial Parasites Infection by usi...

引用

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Meenakshisundaram, N. Sajiv, G. Saveetha University Department of Electronics and Communication Engineering Chennai India

ISBN: (纸本)9798331508456

Malaria remains a significant global health challenge, particularly in resource-limited regions, necessitating accurate and rapid diagnostic tools. This study introduces deepMalariaNet, a deep learning model developed for detecting and classifying malarial parasites using the 'Malaria Parasite image - Malaria Species' dataset from Kaggle. The model employs Residual Attention Mechanisms and Parallel Convolutional Stacks (PCS) to improve diagnostic accuracy by focusing on critical image regions and capturing multi-scale features. Experimental results demonstrate that deepMalariaNet achieves 98.5% accuracy for binary classification (infected vs. non-infected) and 95.2% for multiclass classification (species identification). The model's robustness is validated through 10-fold cross-validation and ablation studies, and it outperforms state-of-the-art models such as ResNet-50 and DenseNet-121 in both accuracy and inference time. deepMalariaNet shows significant promise for real-time malaria detection in clinical settings, contributing to early and accurate diagnosis, which is crucial for effective malaria control and treatment. © 2025 IEEE.

关键词： Photointerpretation

来源：评论

学校读者我要写书评

暂无评论

Seeing Through the Haze: A Comprehensive Review of Underwater image Enhancement Techniques

引用

IEEE ACCESS 2024年 12卷 145206-145233页

作者： Saad Saoud, Lyes Elmezain, Mahmoud Sultan, Atif Heshmat, Mohamed Seneviratne, Lakmal Hussain, Irfan Khalifa Univ Khalifa Univ Ctr Autonomous Robot Syst Abu Dhabi U Arab Emirates

Underwater imaging suffers from significant quality degradation due to light scattering and absorption by water molecules, leading to color cast and reduced visibility. This hinders the ability to analyze and interpret the underwater world. image dehazing techniques have emerged as a crucial component for underwater image enhancement (UIE). This review comprehensively examines both traditional methods, rooted in the physics of light transmission in water, and recent advances in learning-based approaches, particularly deep learning architectures like Convolutional Neural Networks (CNNs), Generative Adversarial Networks (GANs), and Transformers. We conduct a comparative analysis across various metrics, including visual quality, color fidelity, robustness to noise, and computational efficiency, to highlight the strengths and weaknesses of each approach. Furthermore, we address key challenges and future directions for traditional and learning-based methods, focusing on domain adaptation, real-time processing, and integrating physical priors into deep learning models. This review provides valuable insights and recommendations for researchers and practitioners in underwater image enhancement.

关键词： image color analysis image enhancement Sensitivity Reviews Colored noise image restoration Absorption Underwater navigation Underwater image enhancement traditional dehazing methods learning-based dehazing methods deep learning for underwater imaging

来源：评论

学校读者我要写书评

暂无评论

Tomato Plant Leaf Disease Detection using deep learning & image processing

Tomato Plant Leaf Disease Detection using Deep Learning & Im...

引用

International Conference on Signal processing, Communication, Power and Embedded System (SCOPES)

作者： Aniket Chattopadhyay Afreen S Soumya Ranjan Mohapatra Prangya Parimita Pradhan SENSE Vellore Institute of Technology Chennai Tamil Nadu Department of Electrical Engineering CUTM Paralakhemundi Odisha

ISBN: (数字)9798331506452

ISBN: (纸本)9798331506469

This paper presents a systematic analysis for tomato plant leaf disease detection by adapting traditional Convolutional Neural Network (CNN) and pre-trained RESNET50 model. To understand the problem, first a traditional CNN model has been considered. However, the performance of the model was both space and time consuming & provided inadequate predictions. Therefore, to resolve this issue RESNET50 model has been employed. Further, comparing the yielded performance from both CNN and RESNET50, it can be observed RESNET50 exhibits better performance as compared to CNN in terms of accuracy, prediction and complexity.

关键词： Training deep learning Adaptation models Analytical models Accuracy Plants (biology) Feature extraction Convolutional neural networks Residual neural networks Diseases

来源：评论

学校读者我要写书评

暂无评论

deep learning image burst stacking to reconstruct high-resolution ground-based solar observations

引用

ASTRONOMY & ASTROPHYSICS 2025年 693卷

作者： Schirninger, C. Jarolim, R. Veronig, A. M. Kuckein, C. Karl Franzens Univ Graz Inst Phys Univ Pl 5 A-8010 Graz Austria Natl Ctr Atmospher Res High Altitude Observ 3080 Ctr Green Dr Boulder CO USA Karl Franzens Univ Graz Kanzelhohe Observ Solar & Environm Res Graz Austria Inst Astrofis Canarias IAC Via Lactea S-N E-38205 San Cristobal la Laguna Tenerife Spain Univ La Laguna Dept Astrofis E-38206 San Cristobal la Laguna Tenerife Spain Max Planck Inst Sonnensystemforsch Justus Von Liebig Weg 3 D-37077 Gottingen Germany

Context. Large aperture ground-based solar telescopes allow the solar atmosphere to be resolved in unprecedented detail. However, ground-based observations are inherently limited due to Earth's turbulent atmosphere, requiring image correction techniques. Aims. Recent post-image reconstruction techniques are based on using information from bursts of short-exposure images. Shortcomings of such approaches are the limited success, in case of stronger atmospheric seeing conditions, and computational demand. real-time post-image reconstruction is of high importance to enabling automatic processing pipelines and accelerating scientific research. In an attempt to overcome these limitations, we provide a deep learning approach to reconstruct an original image burst into a single high-resolution high-quality image in real time. Methods. We present a novel deep learning tool for image burst reconstruction based on image stacking methods. Here, an image burst of 100 short-exposure observations is reconstructed to obtain a single high-resolution image. Our approach builds on unpaired image-to-image translation. We trained our neural network with seeing degraded image bursts and used speckle reconstructed observations as a reference. With the unpaired image translation, we aim to achieve a better generalization and increased robustness in case of increased image degradations. Results. We demonstrate that our deep learning model has the ability to effectively reconstruct an image burst in real time with an average of 0.5 s of processing time while providing similar results to standard reconstruction methods. We evaluated the results on an independent test set consisting of high- and low-quality speckle reconstructions. Our method shows an improved robustness in terms of perceptual quality, especially when speckle reconstruction methods show artifacts. An evaluation with a varying number of images per burst demonstrates that our method makes efficient use of the combined image info

关键词： atmospheric effects techniques: image processing telescopes Sun: atmosphere Sun: photosphere

来源：评论

学校读者我要写书评

暂无评论

A cognitive deep learning approach for medical image processing

引用

SCIENTIFIC REPORTS 2024年第1期14卷 1-17页

作者： Fakhouri, Hussam N. Alawadi, Sadi Awaysheh, Feras M. Alkhabbas, Fahed Zraqou, Jamal Univ Petra Dept Data Sci & Artificial Intelligence Amman Jordan Blekinge Inst Technol Dept Comp Sci Karlskrona Sweden Univ Santiago De Compostela Comp Graph & Data Engn COGRADE Res Grp Santiago De Compostela Spain Univ Tartu Inst Comp Sci Delta Res Ctr Tartu Estonia Malmo Univ Internet Things & People Res Ctr Malmo Sweden Malmo Univ Dept Comp Sci & Media Technol Malmo Sweden Univ Petra Fac Informat Technol Virtual & Augment Real Dept Amman Jordan

In ophthalmic diagnostics, achieving precise segmentation of retinal blood vessels is a critical yet challenging task, primarily due to the complex nature of retinal images. The intricacies of these images often hinder the accuracy and efficiency of segmentation processes. To overcome these challenges, we introduce the cognitive DL retinal blood vessel segmentation (CoDLRBVS), a novel hybrid model that synergistically combines the deep learning capabilities of the U-Net architecture with a suite of advanced image processing techniques. This model uniquely integrates a preprocessing phase using a matched filter (MF) for feature enhancement and a post-processing phase employing morphological techniques (MT) for refining the segmentation output. Also, the model incorporates multi-scale line detection and scale space methods to enhance its segmentation capabilities. Hence, CoDLRBVS leverages the strengths of these combined approaches within the cognitive computing framework, endowing the system with human-like adaptability and reasoning. This strategic integration enables the model to emphasize blood vessels, accurately segment effectively, and proficiently detect vessels of varying sizes. CoDLRBVS achieves a notable mean accuracy of 96.7%, precision of 96.9%, sensitivity of 99.3%, and specificity of 80.4% across all of the studied datasets, including DRIVE, STARE, HRF, retinal blood vessel and Chase-DB1. CoDLRBVS has been compared with different models, and the resulting metrics surpass the compared models and establish a new benchmark in retinal vessel segmentation. The success of CoDLRBVS underscores its significant potential in advancing medical image processing, particularly in the realm of retinal blood vessel segmentation.

关键词：

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：