检索结果-内蒙古大学图书馆

Multi-source data-driven approach for prediction of melt density during polymer compounding

POLYMER ENGINEERING AND SCIENCE 2024年第6期64卷 2627-2639页

作者： Zhang, Bin-Bin Chen, Zhu-Yun Zhang, Fei Jin, Gang South China Univ Technol Natl Engn Res Ctr Novel Equipment Polymer Proc Key Lab Polymer Proc Engn Minist Educ Guangzhou Peoples R China South China Univ Technol Sch Mech & Automot Engn Guangzhou Peoples R China

Melt density is a crucial quality indicator for polymer composites, yet real-time measurement remains challenging due to processing complexities. While existing machine learning methods offer solutions, they often fall short in complex compounding scenarios. This study presents a novel multi-source data-driven approach for measuring melt density in polycarbonate/acrylonitrile butadiene styrene blends. By incorporating ultrasonic, near-infrared, and Raman spectra data acquired during melt processing, a deep separable convolutional neural network model is developed to predict melt density accurately. The model effectively fuses multi-source data to establish the mapping relationship between input data and melt density output. Results demonstrate the model's ability to monitor melt density in real-time, achieving a prediction accuracy with RMSE and R2 indexes of 0.005 g/cm3 and 0.9841, respectively. The proposed approach outperforms existing methods, showcasing its effectiveness and superiority in melt density prediction for polymer compounding *** Establishment of the real-time monitoring system for polymer extrusion processes. Conversion of multi-sensor signals into time-frequency images using wavelet decomposition. Fusion of sensor data into a three-channel tensor-image. Development of a data-driven DSCNN model for predicting melt density. Implementation for online monitoring and prediction in PC/ABS compounding system. Sensor data acquisition and DSCNN model monitoring process diagram. image

关键词： data fusion data-driven density measurement DSCNN extrusion processing in-line monitoring polymer melt

来源：评论

学校读者我要写书评

暂无评论

deep Prior-Based Audio Inpainting Using Multi-Resolution Harmonic Convolutional Neural Networks

引用

IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE processing 2024年 32卷 113-123页

作者： Miotello, Federico Pezzoli, Mirco Comanducci, Luca Antonacci, Fabio Sarti, Augusto Politecn Milan Dipartimento Elettron Informaz & Bioingn I-20133 Milan Italy

In this manuscript, we propose a novel method to perform audio inpainting, i.e., the restoration of audio signals presenting multiple missing parts. Audio inpainting can be interpreted in the context of inverse problems as the task of reconstructing an audio signal from its corrupted observation. For this reason, our method is based on a deep prior approach, a recently proposed technique that proved to be effective in the solution of many inverse problems, among which image inpainting. deep prior allows one to consider the structure of a neural network as an implicit prior and to adopt it as a regularizer. Differently from the classical deep learning paradigm, deep prior performs a single-element training and thus it can be applied to corrupted audio signals independently from the available training data sets. In the context of audio inpainting, a network presenting relevant audio priors will possibly generate a restored version of an audio signal, only provided with its corrupted observation. Our method exploits a time-frequency representation of audio signals and makes use of a multi-resolution convolutional autoencoder, that has been enhanced to perform the harmonic convolution operation. Results show that the proposed technique is able to provide a coherent and meaningful reconstruction of the corrupted audio. It is also able to outperform the methods considered for comparison, in its domain of application.

关键词： time-frequency analysis Task analysis deep learning Inverse problems Convolution Spectrogram image restoration Audio inpainting deep prior deep learning

来源：评论

学校读者我要写书评

暂无评论

FruitQ: a new dataset of multiple fruit images for freshness evaluation

引用

MULtimeDIA TOOLS AND APPLICATIONS 2024年第4期83卷 11433-11460页

作者： Abayomi-Alli, Olusola O. O. Damasevicius, Robertas Misra, Sanjay Abayomi-Alli, Adebayo Kaunas Univ Technol Dept Software Engn LT-44249 Kaunas Lithuania Inst Energy Technol Dept Appl Data Sci N-1777 Halden Norway Fed Univ Agr Dept Comp Sci Abeokuta 110124 Nigeria

Application of artificial intelligence methods in agriculture is gaining research attention with focus on improving planting, harvesting, post-harvesting, etc. Fruit quality recognition is crucial for farmers during harvesting and sorting, for food retailers for quality monitoring, and for consumers for freshness evaluation, etc. However, there is a lack of multi-fruit datasets to support real-time fruit quality evaluation. To address this gap, we present a new dataset of fruit images aimed at evaluating fruit freshness, which addresses the lack of multi-fruit datasets for real-time fruit quality evaluation. The dataset contains images of 11 fruits categorized into three freshness classes, and five well-known deep learning models (ShuffleNet, SqueezeNet, EfficientNet, ResNet18, and MobileNet-V2) were adopted as baseline models for fruit quality recognition using the dataset. The study provides a benchmark dataset for the classification task, which could improve research endeavors in the field of fruit quality recognition. The dataset is systematically organized and annotated, making it suitable for testing the performance of state-of-the-art methods and new learning classifiers. The research community in the fields of computer vision, machine learning, and pattern recognition could benefit from this dataset by applying it to various research tasks such as fruit classification and fruit quality recognition. The study achieved impressive results with the best classifier being ResNet-18 with an overall best performance of 99.8% for accuracy. The study also identified limitations, such as the small size of the dataset, and proposed future work to improve deep learning techniques for fruit quality classification tasks.

关键词： fruit freshness evaluation fruit decay detection precision agriculture image processing computer vision

来源：评论

学校读者我要写书评

暂无评论

An Overview of Moving Object Detection Using YOLO deep learning Models 2

An Overview of Moving Object Detection Using YOLO Deep Learn...

引用

2nd International Conference on Disruptive Technologies, ICDT 2024

作者： Dwivedi, Upendra Joshi, Kireet Shukla, Surendra Kumar Rajawat, Anand Singh G L Bajaj Institute of Technology and Management Department of Applied Computational Science and Engineering Greater Noida India Graphic Era Deemed to Be University Department of Computer Science and Engineering Dehradun India SVKM S Nmims Mpstme Department of Computer Science and Engineering Shirpur Campus Shirpur India School of Computer Sciences and Engineering Sandip University Nashik India

ISBN: (纸本)9798350371055

Computer vision is a promising domain that focuses on emerging approaches, algorithms and technologies to provide computing capability to machine to analysis visual data, such as image files, videos files and real time video streaming. In Computer Vision and image processing detecting object from images and videos has been topic of extensive research. This is accomplished by applying various computer vision techniques to analyze the visual data and determine class of objects from image and videos files. One widespread approach to object detection is using the deep learning models. YOLO (You Only Look Once) is a convolutional neural network that offers a fast and efficient solution to object detection using deep learning. It enable computer to handle real-time detection of objects in video frames and accurately locate and classify moving objects. YOLO is able to simultaneously detect and classify objects in an efficient manner using convolutional neural networks. © 2024 IEEE.

关键词： deep learning image processing Motion Detection Moving Object Detection YOLO

来源：评论

学校读者我要写书评

暂无评论

A real-time deep learning network for ship detection in SAR images

引用

SIGNAL image AND VIDEO processing 2024年第2期18卷 1893-1899页

作者： Zhou, Wenxue Zhang, Huachun Chinese Acad Sci Aerosp Informat Res Inst Beijing 100094 Peoples R China Univ Chinese Acad Sci Sch Elect Elect & Commun Engn Beijing 100049 Peoples R China

In the field of ship detection in synthetic aperture radar (SAR) images using deep learning, traditional models often face challenges related to their complex structure and high computational demands. To address these problems, this paper proposes a lightweight end-to-end convolutional neural network called BGD-Net, which is based on anchor-free methods. This network incorporates a novel feature pyramid called Ghost, Cross-Stage-Partial, and Path Aggregation Network (G-CSP-PAN) to enhance the detection performance across targets of varying scales. Additionally, it introduces an efficient decoupled detection head, termed the efficient decoupled head (ED-Head), to enhance the interaction between regression and classification. Furthermore, an optimized loss function named optimized efficient intersection over union (OEIoU) loss is proposed for edge regression. The proposed method is evaluated on public datasets, SSDD and SAR-Ship-Dataset, demonstrating a balance between detection accuracy and efficiency.

关键词： Anchor-free One-stage detection Synthetic aperture radar Ship detection Decoupled head

来源：评论

学校读者我要写书评

暂无评论

deep optimal feature extraction and selection-based motor fault diagnosis using vibration

引用

ELECTRICAL ENGINEERING 2024年第5期106卷 6339-6358页

作者： Jigyasu, Rajvardhan Shrivastava, Vivek Singh, Sachin Natl Inst Technol Delhi Dept Elect Engn Delhi 110036 India

The rolling elements of the induction motor are highly susceptible to faults. The detection and diagnosis of rolling element faults are accurate and reliable only when the extracted features are accurate. The paper proposes an approach for bearing and rotor fault diagnosis using deep optimal feature extraction and selection based on vibration signal analysis. The deep feature extraction is done using an ensemble deep models features extraction approach in which features are extracted from seven pretrained models are fused serially using serial-based feature fusion technique. This leads to a solution for a higher efficacy model, but at the cost of high processing time as the feature data set gets large. A unique approach termed Ensemble Feature Selection has been developed to address this issue and limit the harmful impact of unwanted features in data-driven diagnostics. The processing time is further reduced using the shallow classifier at the fully connected layer. The proposed model is tested using the data acquired in the laboratory and validated using the available online benchmark data sets.

关键词： Motor fault diagnosis Vibration analysis deep learning Feature extraction Feature selection Signal processing image processing

来源：评论

学校读者我要写书评

暂无评论

Enhancing Digit Recognition for Luminous images in Edge Computing Through Transfer learning With Robustness and Fault Tolerance

引用

IEEE TRANSACTIONS ON RELIABILITY 2024年第2期74卷 2529-2537页

作者： Hsu, Tse-Chuan Tsai, Yao-Hong Chu, William Cheng-Chung Soochow Univ Dept Comp Sci & Informat Management Taipei 10048 Taiwan Hsuan Chuang Univ Dept Visual Commun Hsinchu 30092 Taiwan Tunghai Univ Dept Comp Sci Tiachung 407705 Taiwan

deep learning is developing rapidly, and the emergence of many network architectures has brought significant breakthroughs to training recognition models. Due to the maturity of edge computing technology, we can perform regional image training through distributed nodes, which significantly improves the training model's accuracy while performing transfer learning to achieve better performance. In image processing technology, high-precision recognition of non-luminous images can currently be achieved by modeling, if we replace the visual recognition target with a glowing digital panel, the recognition rate cannot be the same as the static text recognition rate. This article uses Keras to build a convolutional neural networks deep learning model to identify glowing light-emitting diodes (LED) digits, incremental learning to complete transfer learning on edge computing nodes, and an integrated IoT architecture to achieve better recognition results. In the experiment, the verification results obtained from the distributed training nodes were successfully combined to model and retrain the nodes. The proposed distributed learning method can increase the accuracy from 70% to 89%. At the same time, the misclassified images can be retrained by integrating the transfer learning model with the distributed learning results, and the accuracy reaches more than 92%.

关键词： image recognition Training Computational modeling Edge computing Data models Transfer learning Text recognition deep learning digit recognition incremental learning Keras transfer learning

来源：评论

学校读者我要写书评

暂无评论

FireNet: A Hybrid deep learning Approach for Enhanced Fire Detection in Remote Sensing imagery

FireNet: A Hybrid Deep Learning Approach for Enhanced Fire D...

引用

Intelligent Systems Conference

作者： Bukhari, Syed Muhammad Salman Zafar, Muhammad Hamza Moosavi, Syed Kumayl Raza Khan, Noman Mujeeb Sanfilippo, Filippo Capital Univ Sci & Technol Dept Elect Engn Islamabad Pakistan Univ Agder Dept Engn Sci Grimstad Norway Natl Univ Sci & Technol SEECS Islamabad Pakistan Kaunas Univ Technol Dept Software Engn Kaunas Lithuania

ISBN: (纸本)9783031663284;9783031663291

In this study, we address the critical task of early bush-fire detection in remote areas through UAV imaging, a vital effort for protecting inhabitants, infrastructure, and ecosystems within the smart cities paradigm. We introduce FireNet, a deep learning framework that employs a hybrid Inception Residual Neural Network (Inception-ResNet) model, specifically designed for the rapid and precise classification of fire and non-fire regions in real-time image processing scenarios. FireNet synergises the strengths of both Inception and ResNet architectures, enhancing the model's feature extraction capabilities while significantly reducing computational overhead. A distinctive aspect of FireNet is the adoption of the HardSwish activation function, demonstrating superior performance over the conventional Rectified Linear Unit (ReLU) in our fire detection cases. Through rigorous evaluation of a robust dataset, FireNet achieves an impressive accuracy of 96%, with a corresponding AUC of approximately 96%. These results not only affirm FireNet's efficacy in accurately identifying fire and non-fire areas but also underscore its importance as a crucial tool for real-time fire monitoring and rescue operations, especially in the context of smart cities. With its outstanding efficiency and accuracy, FireNet represents a significant leap forward in the domain of fire detection technologies, highlighting the role of deep learning in enhancing urban resilience against fires.

关键词： deep learning Fire detection UAV imaging Inception-ResNet hybrid model real-time image processing Smart cities

来源：评论

学校读者我要写书评

暂无评论

New Statistic Detector for Structural image Similarity

引用

IEEE TRANSACTIONS ON SIGNAL processing 2025年 73卷 1168-1183页

作者： Diaw, Moustapha Retraint, Florent Morain-Nicolier, Frederic Delahaies, Agnes Landre, Jerome Univ Reims CReST EA 3804 IUT Troyes F-51097 Reims France Univ Technol Troyes LIST3N UTT F-10300 Troyes France Jonkoping Univ Comp Sci & Informat Dept S-55318 Jonkoping Sweden

Social networks like LinkedIn, Facebook, and Instagram contribute significantly to the rise of image prevalence in daily life, with numerous images posted in everyday. Detecting image similarity is crucial for many applications. While deep learning methods like Learned Perceptual image Patch Similarity (LPIPS) are popular, they often overlook image structure. An alternative method involves using pre-trained models ($e.g.$, LeNet-$5$ and VGG-$16$) to extract features and employing classifiers. However, deep learning methods demand substantial computational resources and they also suffer from uncontrolled false alarms. This paper proposes a novel Generalized Likelihood Ratio Test (GLRT) detector based on a hypothesis testing framework to identify the similarity of structural image pairs. The proposed approach minimizes the need for extensive computational resources, and false alarms can be regulated by employing a threshold. The detector is applied to Local Dissimilarity Maps (LDM), with gray-level values modeled by a statistical distribution. Experimental results on simulated and real data confirm its effectiveness for structural similarity detection. Additionally, a Simple Likelihood Ratio Test (SLRT) is tested on simulated data. Comparisons with deep learning and classical measures like Structural Similarity Index (SSIM) and Feature Similarity Index (FSIM) show the proposed detector performs comparably or better in terms of Area Under the Curve (AUC) with less computing time, especially for structural similarity.

关键词： Feature extraction Detectors Convolutional neural networks Transformers image quality Noise Computational modeling Indexes image classification Training Classification local dissimilarity map weibull distribution GLRT similarity measures pre-trained CNN

来源：评论

学校读者我要写书评

暂无评论

An Efficient Approach to Escalate the Speed of Training Convolution Neural Networks

引用

China Communications 2024年第2期21卷 258-269页

作者： P Pabitha Anusha Jayasimhan Department of Computer Technology Anna UniversityMadras Institute of Technology CampusChennai 600044India

deep neural networks excel at image identification and computer vision applications such as visual product search, facial recognition, medical image analysis, object detection, semantic segmentation,instance segmentation, and many others. In image and video recognition applications, convolutional neural networks(CNNs) are widely employed. These networks provide better performance but at a higher cost of computation. With the advent of big data, the growing scale of datasets has made processing and model training a time-consuming operation, resulting in longer training times. Moreover, these large scale datasets contain redundant data points that have minimum impact on the final outcome of the model. To address these issues, an accelerated CNN system is proposed for speeding up training by eliminating the noncritical data points during training alongwith a model compression method. Furthermore, the identification of the critical input data is performed by aggregating the data points at two levels of granularity which are used for evaluating the impact on the model *** experiments are conducted using the proposed method on CIFAR-10 dataset on ResNet models giving a 40% reduction in number of FLOPs with a degradation of just 0.11% accuracy.

关键词： CNN deep learning image classification model compression

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：