检索结果-内蒙古大学图书馆

A Survey on Computer vision Architectures for Large Scale image Classification using Deep Learning

INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND applications 2021年第10期12卷 105-120页

作者： Himabindu, D. Dakshayani Kumar, S. Praveen VNRVJIET Dept IT Hyderabad 90 TS India GITAM Dept Comp Sci Visakhapatnam 45 Andhra Pradesh India

The advancement in deep learning is increasing day-by-day from image classification to language understanding tasks. In particular, the convolution neural networks are revived and shown their performance in multiple fields such as natural language understanding, signal processing, and computer vision. The property of translational invariance for convolutions has made a huge advantage in the field of computer vision to extract feature invariances appropriately. When these convolutions trained using back-propagation tend to prove their results ability to outperform existing machine vision techniques by overcoming the various hand-engineered machine vision models. Hence, a clear understanding of current deep learning methods is crucial. These convolution neural networks have proven to show their performance by attaining state-of-the-art performance in computer vision over years when applied on humongous data. Hence in this survey, we detail a set of state-of-the-art models in image classification evolved from the birth of convolutions to present ongoing research. Each state-of-the-art model evolved in the successive year is illustrated with architecture schema, implementation details, parametric tuning and their performance. It is observed that the neural architecture construction i.e. a supervised approach for an image classification problem is evolved as data construction with cautious augmentations i.e., a self-supervised approach. A detailed evolution from neural architecture construction to augmentation construction is illustrated by provided appropriate suggestions to improve the performance. Additionally, the implementation details and the appropriate source for the execution and reproducibility of results are tabulated.

关键词： image classification deep learning computer vision survey convolution neural networks imageNET dataset

来源：评论

学校读者我要写书评

暂无评论

Multiple Activation Functions and Data Augmentation-Based Lightweight Network or In Situ Tool Condition Monitoring

引用

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS 2022年第12期69卷 13656-13664页

作者： You, Zhichao Gao, Hongli Li, Shichao Guo, Liang Liu, Yuekai Li, Jingbo Southwest Jiaotong Univ Engn Res Ctr Adv Driving Energy Saving Technol Sch Mech Engn Minist Educ Chengdu 610031 Peoples R China

Intelligent manufacturing raises higher requirements for tool condition monitoring (TCM) in terms of accuracy, robustness, and adaptability. At present, direct methods based on image processing and deep learning have made breakthroughs in TCM. However, some issues, such as image quality, model parameters, and dataset scale in the abovementioned methods, restrict industrial applications of TCM. Regarding the abovementioned issue, the purpose of this article is to propose a lightweight network model based on multiple activation functions to promote the intelligent industrial application of TCM. First, the image quality mechanism caused by complex working conditions is analyzed in industrial environments. Correspondingly, data augmentation is adopted to solve the problem of data scale under the premise of ensuring data quality and richness. Then, the adaptive activation function and the hard version of swish are introduced at the front and second half of the network to avoid information loss and reduce the activation function cost. Finally, a lightweight network based on cloud-edge collaboration for TCM is constructed. The model is iteratively optimized in the cloud and inferenced on the edge embedded device. The accuracy and adaptability of the proposed network are verified by accelerating milling cutter life under multiple working conditions.

关键词： Cloud-edge collaboration data augmentation lightweight network machine vision multiple activation functions tool condition monitoring (TCM)

来源：评论

学校读者我要写书评

暂无评论

Current trends and future orientation in diagnosing lung pathologies:A systematic survey

引用

Intelligent Medicine 2025年第1期5卷 23-36页

作者： Tamim M.Al-Hasan Mohammad Noorizadeh Faycal Bensaali Nader Meskin Ali Ait Hssain Department of Electrical Engineering College of EngineeringQatar UniversityDohaP.O.Box 2713Qatar Medical Intensive Care Unit Hamad Medical CorporationDohaQatar

Lung diseases pose a significant threat to public health worldwide,resulting in a substantial number of *** such as chronic obstructive pulmonary disease and lung cancer constitute two of the three deadliest diseases worldwide,contributing to over 3 million deaths *** study offered a comparative analysis of different diagnostic techniques used for lung pathologies from an engineering *** review concentrated on intelligent detection methods,including electronic nose,computer vision(CV),or image processing,and biosensors such as graphene-field effect transistor(FET).The E-nose-based detection technique uses electronic sensors to recognize volatile organic compounds(VOCs)in the exhaled *** VOCs can aid in the diagnosis of lung pathologies such as *** CV processing method involves the application of advanced imaging techniques and machine learning algorithms to scrutinize and diagnose lung pathologies and ventilatorassociated pneumonia(VAP).Lastly,biosensors employ the exceptional properties of these materials to identify specific biomarkers in biological *** information can be used to diagnose lung pathologies and *** study examined the current state-of-the-art methods and offers a comprehensive analysis of their advantages and disadvantages from an engineering *** study underscored the potential of these techniques to enhance the diagnosis of lung pathologies and VAP and presents the advances in the field of smart biomedical ***,it emphasized the necessity for further research to optimize their performance and clinical usefulness.

关键词： Biosensors Bio-signal processing Electronic nose Lung disease Nosocomial infections Ventilator-associated pneumonia

来源：评论

学校读者我要写书评

暂无评论

vision-based modal analysis of cutting tools

引用

CIRP JOURNAL OF MANUFACTURING SCIENCE AND TECHNOLOGY 2021年 32卷 91-107页

作者： Gupta, Pulkit Rajput, Harsh Singh Law, Mohit Indian Inst Technol Kanpur Dept Mech Engn Machine Tool Dynam Lab Kanpur 208016 Uttar Pradesh India

This paper presents the use of vision-based methods for cutting tool motion registration and modal analysis. Motion of three illustrative tools were recorded using low- and high-speed cameras with sufficiently high resolutions. The tool's own features are used to register motion. Pixels within images from recordings of the vibrating tools are treated as non-contact motion sensors. Comparative analysis of three different methods of motion registration are presented to evaluate their suitability for the application of interest. These include variants of expanded edge detection and tracking schemes, expanded optical flow-based schemes, and established digital image correlation methods. Performance of different methods was observed to be governed by the tool's own features, illumination conditions, noise, and the image acquisition parameters. Extracted motion was benchmarked against twice integrated measured tool point accelerations, and motion was generally observed to compare well. Modal parameters extracted from vision-based measurements were also observed to agree with those extracted using more traditional experimental modal analysis procedures using a contact type accelerometer as the transducer. Since methods presented are generalized, they can suitably be adapted for other applications of interest. (C) 2020 CIRP.

关键词： Dynamics Cutting tool Computer vision Vibration image processing Digital image correlation

来源：评论

学校读者我要写书评

暂无评论

Assessment of 3D MRI image segmentation and Classification for Brain Tumor Detection Using ConvLSTM 5

Assessment of 3D MRI Image segmentation and Classification f...

引用

5th IEEE International Conference on Cybernetics, Cognition and machine Learning applications, ICCCMLA 2023

作者： Raju, K Srujan Arvind, Sudha Chegoni, Ramesh Naryana, V.A. Vivekananda, A. Kishore Babu, Ch Raja CMR Technical Campus Department of CSE Telangana Hyderabad India CMR Technical Campus Department of ECE Telangana Hyderabad India CMR Technical Campus Department of IT Telangana Hyderabad India CMR College of Engineering & Technology Department of CSE Telangana Hyderabad India

ISBN: (纸本)9798350338287

The human brain serves as the principal controller of the humanoid system. Brain tumors are the result of abnormal cell division and proliferation, and the development of these tumors can result in brain cancer. The use of computer vision in diagnostic procedures has the potential to lessen human mistake in judgment. The incorporation of new technology in healthcare is seen as a technique to improve human decision-making in the area of diagnosis. Magnetic Resonance Imaging (MRI) is thought to be comparatively more dependable and secure than other diagnostic imaging techniques. In order to identify brain tumors on (BraTS), we suggested a method using Convolutional Long Short-Term Memory (ConvLSTM) on segmented anomalous portions of 3D MRI brain images in Matlab. Using Matlab, a graphical user interface that is simple to use is created to find brain tumors. This effort sought to identify the precise tumor site by first classifying the findings from various brains imaging into three categories: normal, benign, and ***- Deep Learning, Pytorch, Neural Network, Artificial Intelligence, Natural Language processing, Tkinter. © 2023 IEEE.

关键词： Magnetic resonance imaging

来源：评论

学校读者我要写书评

暂无评论

Receptive Field-Based All-Optical Spiking Neural Network for image processing

引用

IEEE JOURNAL OF QUANTUM ELECTRONICS 2024年第1期60卷 1页

作者： Chen, Taiyi Huang, Yu Zhou, Pei Mu, Penghua Xiang, Shuiying Chizhevsky, V. N. Li, Nianqiang Soochow Univ Collaborat Innovat Ctr Suzhou Nano Sci & Technol Key Lab Adv Opt Mfg Technol Jiangsu Prov Sch Optoelect Sci & Engn Suzhou 215006 Peoples R China Soochow Univ Key Lab Modern Opt Technol Educ Minist China Suzhou 215006 Peoples R China Yantai Univ Inst Sci & Technol Optoelect Informat Yantai 264005 Peoples R China Xidian Univ State Key Lab Integrated Serv Networks Xian 710071 Peoples R China Natl Acad Sci Belarus BI Stepanov Inst Phys Minsk 220072 BELARUS

We report on a novel structure of a receptive field (RF)-based multi-layer all-optical neural network using a micropillar laser with a saturable absorber (SA) for image processing. From the perspective of biological vision, the realization of image processing based on the RF provides the biological rationality for the machine vision implemented by the spiking neural network (SNN). By exploiting the fast physical mechanisms of gain and absorption in the SA laser, the photonic spike-timing-dependent plasticity (STDP) curves are achieved to train the weights. Here, the source image pixels are mapped into the temporal information of spike trains injected into the neural network through the temporal coding method called time-to-first-spike encoding. Different source images are processed and tested by the proposed photonic SNN. Simulation results show that our proposed system can process not only simple binary images but also complex color images under the adjustment of STDP rules. When considering the robustness, we demonstrate the tolerance of the image segmentation to the time jitter. These results indicate that our proposed photonic SNN can achieve high-resolution processing of complex source images. Additionally, the time-multiplexing technique can be further adopted to simplify the RF structure, which is expected to reduce the complexity of the whole system, thus facilitating physical applications. Our work offers the prospect for a high-speed photonic spiking platform for image processing.

关键词： Neurons Photonics Biological information theory Radio frequency Optical imaging Laser excitation Biomedical optical imaging Photonic spiking neural networks receptive field (RF) photonic neural networks neuromorphic photonics excitable lasers spike-timing-dependent plasticity (STDP) time-to-first-spike (TTFS)

来源：评论

学校读者我要写书评

暂无评论

The Need for machines for the Nondestructive Quality Assessment of Potatoes with the Use of Artificial Intelligence Methods and Imaging Techniques

引用

SENSORS 2023年第4期23卷 1787页

作者： Danielak, Marek Przybyl, Krzysztof Koszela, Krzysztof Poznan Univ Life Sci Dept Biosyst Engn Wojska Polskiego 50 PL-60625 Poznan Poland Poznan Inst Technol Lukasiewicz Res Network Starolecka 31 PL-60963 Poznan Poland Poznan Univ Life Sci Fac Food Sci & Nutr Dept Dairy & Proc Engn Wojska Polskiego 31 PL-60624 Poznan Poland

This article describes chemical and physical parameters, including their role in the storage, trade, and processing of potatoes, as well as their nutritional properties and health benefits resulting from their consumption. An analysis of the share of losses occurring during the production process is presented. The methods and applications used in recent years to estimate the physical and chemical parameters of potatoes during their storage and processing, which determine the quality of potatoes, are presented. The potential of the technologies used to classify the quality of potatoes, mechanical and ultrasonic, and image processing and analysis using vision systems, as well as their use in applications with artificial intelligence, are discussed.

关键词： potato tuber quality assessment sorting machines grading ultrasound neural networks machine vision nondestructive method noninvasive method

来源：评论

学校读者我要写书评

暂无评论

Incremental Learning Model Based on Ensemble Learning

Incremental Learning Model Based on Ensemble Learning

引用

image processing, Computer vision and machine Learning (ICICML), International Conference on

作者： Tong Zhang Weiqiang Wu College of Mechanical and Electrical Engineering Guilin University of Electronic Technology Guilin China

ISBN: (数字)9798350355413

ISBN: (纸本)9798350355420

Currently, deep learning neural networks have shown remarkable performance in the field of image classification. However, existing classification networks are typically applied within closed environments, where all collected samples are batch-trained to produce classification results. In practical applications, the categories of samples are not static but dynamically increase over time. With the continual addition of new classes, the current classification methods exhibit limited continual learning capabilities, making them unsuitable for applications with stringent real-time requirements. To address this issue, this paper investigates an incremental learning algorithm based on ensemble transfer learning to handle classification training for newly added categories. This algorithm leverages the strengths of ensemble transfer learning to enhance the network's ability to continually learn new categories, enabling fast and accurate classification of new samples. In this paper, the proposed algorithm is applied to the CUB-200-2011 dataset for class-incremental experiments and compared with other deep learning models. Experimental results demonstrate that the incremental learning classification algorithm based on integrated transfer outperforms other models.

关键词： Deep learning Training Incremental learning machine learning algorithms Accuracy Computational modeling Transfer learning Neural networks Classification algorithms image classification

来源：评论

学校读者我要写书评

暂无评论

Deep Learning, machine Learning - Digital Signal and image processing: From Theory to Application

arXiv

引用

arXiv 2024年

作者： Hsieh, Weiche Bi, Ziqian Liu, Junyu Peng, Benji Zhang, Sen Pan, Xuanhe Xu, Jiawei Wang, Jinlang Chen, Keyu Yin, Caitlyn Heqi Feng, Pohsun Wen, Yizhu Wang, Tianyang Li, Ming Ren, Jintao Niu, Qian Chen, Silin Liu, Ming National Tsing Hua University Taiwan Indiana University United States Kyoto University Japan AppCubic Rutgers University United States University of Wisconsin-Madison United States Purdue University United States Georgia Institute of Technology United States National Taiwan Normal University Taiwan University of Hawaii United States Xi’an Jiaotong-Liverpool University China Aarhus University Denmark Zhejiang University China

Digital Signal processing (DSP) and Digital image processing (DIP) with machine Learning (ML) and Deep Learning (DL) are popular research areas in Computer vision and related fields. We highlight transformative applications in image enhancement, filtering techniques, and pattern recognition. By integrating frameworks like the Discrete Fourier Transform (DFT), Z-Transform, and Fourier Transform methods, we enable robust data manipulation and feature extraction essential for AI-driven tasks. Using Python, we implement algorithms that optimize real-time data processing, forming a foundation for scalable, high-performance solutions in computer vision. This work illustrates the potential of ML and DL to advance DSP and DIP methodologies, contributing to artificial intelligence, automated feature extraction, and applications across diverse domains. Copyright © 2024, The Authors. All rights reserved.

关键词： Discrete Fourier transforms

来源：评论

学校读者我要写书评

暂无评论

On-Line Monitoring System of Welding Quality Based on machine vision and machine Learning

On-Line Monitoring System of Welding Quality Based on Machin...

引用

image processing and Computer applications (ICIPCA), IEEE International Conference on

作者： Genchen Peng Liping Zhang Zhijun Lu Yong Shi Jiangsu XCMG Construction Machinery Research Institute Co. Ltd Xuzhou China Road Machinery Branch XCMG Construction Machinery Co. Ltd Xuzhou China

In this paper, an online monitoring system of welding quality based on machine vision and machine learning was proposed. A high-speed CCD camera was used to monitor the tail end of the molten pool, and the remove small objects algorithm and contour compensation based on convex hull algorithm were utilized to achieve high-precision collection of features such as the width and length of the tail of the molten pool. This effectively solved the technical challenges caused by welding splashes and plasma arc, which could interfere with visual acquisition. Combined with neural network algorithms, a welding quality model was established and validated to accurately identify defects such as welding undercut, welding deviations, and unstable welding processes, with a defect recognition rate of $\geq 94\%$ .

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：