检索结果-南通市图书馆

Deep Learning Based Underground Sewer defect classification Using a Modified RegNet

Computers, Materials & Continua 2023年第6期75卷 5451-5469页

作者： Yu Chen Sagar A.S.M.Sharifuzzaman Hangxiang Wang Yanfen Li L.Minh Dang Hyoung-Kyu Song Hyeonjoon Moon Department of Computer Science and Engineering Sejong UniversitySeoul05006Korea Department of Intelligent Mechatronics Engineering Sejong UniversitySeoul05006Korea Department of Information of Communication Engineering and Convergence Engineering for Intelligent DroneSejong UniversitySeoul05006Korea

The sewer system plays an important role in protecting rainfall and treating urban wastewater.Due to the harsh internal environment and complex structure of the sewer,it is difficult to monitor the sewer system.Researchers are developing different methods,such as the Internet of Things and Artificial Intelligence,to monitor and detect the faults in the sewer system.Deep learning is a promising artificial intelligence technology that can effectively identify and classify different sewer system defects.However,the existing deep learning based solution does not provide high accuracy prediction and the defect class considered for classification is very small,which can affect the robustness of the model in the constraint environment.As a result,this paper proposes a sewer condition monitoring framework based on deep learning,which can effectively detect and evaluate defects in sewer pipelines with high accuracy.We also introduce a large dataset of sewer defects with 20 different defect classes found in the sewer pipeline.This study modified the original RegNet model by modifying the squeeze excitation(SE)block and adding the dropout layer and Leaky Rectified Linear Units(LeakyReLU)activation function in the Block structure of RegNet model.This study explored different deep learning methods such as RegNet,ResNet50,very deep convolutional networks(VGG),and GoogleNet to train on the sewer defect dataset.The experimental results indicate that the proposed system framework based on the modified-RegNet(RegNet+)model achieves the highest accuracy of 99.5 compared with the commonly used deep learning models.The proposed model provides a robust deep learning model that can effectively classify 20 different sewer defects and be utilized in real-world sewer condition monitoring applications.

关键词： Deep learning defect classification underground sewer computer vision convolutional neural network RegNet

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

defect detection and classification for mobile phone cover glass based on visual perception

Defect detection and classification for mobile phone cover g...

引用

第32届中国控制与决策会议

作者： Mingbo Yang Bohao Zhang Jiabin Zhang Hu Su Mechatronics Department North China University of Technology University Institute of Automation Chinese Academy of Sciences University of Chinese Academy of Sciences

In order to improve the product quality,defect detection is a significant concern in the production of MPCG（Mobile Phone Cover Glass）.Aiming at this problem,an automatic defect detection system is developed in this paper.The system adopts backlight imaging strategy to improve the Signal-to-noise ratio（SNR）.With the captured images,an adaptive binarization algorithm is firstly adopted and improved which is based on integral graph to achieve high real-time performance.A multi-dimensional feature vector is then designed for defect classification to eliminate noise interference.Finally,the effectiveness of the proposed algorithm is experimentally verified during MPCG production process.It is further demonstrated that the algorithm and system developed in this paper could satisfy the requirements of both speed and accuracy under the situation of industrial scene.

关键词： MPCG visual perception defect detection defect classification

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Automated defect classification in a honeycombstructure by infraed thermography

Automated defect classification in a honeycombstructure by i...

引用

2018年全国固体力学学术会议

作者： Caiqi Hu Hai Zhang Shicai Liu Jianqiao Meng Yuxia Duan School of Physics and Electronics Central South University Department of Mechanical and Industrial Engineering University of Toronto Department of Electrical and Computer Engineering Universite Laval

In this study, in order to prove our concept of automated defect classification, a flat-bottom holed stainless-steel plate and an aerospace honeycomb structured specimen were manufactured to simulated the air, oil and water ingress. Raw data, and thermographic signal reconstruction(TSR) coefficients were used to train, and test the two constructed, multilayer, feed-forward neural network(NN) models. The testing results show the great potential of industrial application using machine learning to classify defects. The research was then extended to the classification of oil and water ingress, excess glue, and delamination in honeycomb-structured materials.

关键词： defect classification infrared thermography neural networks coefficient honeycomb

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

Algorithmic Scheme for Concurrent Detection and classification of Printed Circuit Board defects

引用

Computers, Materials & Continua 2022年第4期71卷 355-367页

作者： Jakkrit Onshaunjit Jakkree Srinonchat Faculty of Engineering Rajamangala University of Technology Thanyaburi(Main Campus)Khlong LuangPathum Thani12110Thailand

An ideal printed circuit board(PCB)defect inspection system can detect defects and classify PCB defect types.Existing defect inspection technologies can identify defects but fail to classify all PCB defect types.This research thus proposes an algorithmic scheme that can detect and categorize all 14-known PCB defect types.In the proposed algorithmic scheme,fuzzy cmeans clustering is used for image segmentation via image subtraction prior to defect detection.Arithmetic and logic operations,the circle hough transform(CHT),morphological reconstruction(MR),and connected component labeling(CCL)are used in defect classification.The algorithmic scheme achieves 100%defect detection and 99.05%defect classification accuracies.The novelty of this research lies in the concurrent use of CHT,MR,and CCL algorithms to accurately detect and classify all 14-known PCB defect types and determine the defect characteristics such as the location,area,and nature of defects.This information is helpful in electronic parts manufacturing for finding the root causes of PCB defects and appropriately adjusting the manufacturing process.Moreover,the algorithmic scheme can be integrated into machine vision to streamline the manufacturing process,improve the PCB quality,and lower the production cost.

关键词： PCB inspection PCB defect types defect detection defect classification image processing

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

classification algorithms of strip surface defects based on PCA-ELM

Classification algorithms of strip surface defects based on ...

引用

第三十九届中国控制会议

作者： Fengze Miao Ying Tian School of Computer Science and Software Engineering University of Science and Technology Liaoning

The computer image technology has achieved impressive recognition rates in image classification tasks recently.In order to tackle the complex defect features in strip steel,we conduct a study of the effectiveness of principal component analysis and extreme learning machine（ELM） in the classification of strip surface in this paper.Using the principal component analysis,it extracts the characteristics of dimension reduction,obtains the characteristic quantity,reduces the dimension used as the input sample for the Extreme Learning Machine and trains the support vector machine and realizes the recognition.Compared with traditional extreme learning machine and support vector machine（SVM）,PCA-ELM algorithm shows a higher accuracy and better generalization ability,which play a guiding sole on actual production.

关键词： Strip steel surface defect Principal component analysis Extreme learning machine defect classification

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

RD-GRF for Automatic classification of Surface defects of Lithium-ion Battery Electrodes

RD-GRF for Automatic Classification of Surface Defects of Li...

引用

第40届中国控制会议

作者： Junyi Ni Yilin Wu Jiaming Xu Yu Liu School of Automation Science and Engineering South China University of Technology School of Computer Science Guangdong University of Education

Considering that defect classification is an indispensable element in the production process of lithium-ion battery(LIB), a two-step classification method named RD-GRF is proposed in this paper to classify the six types of surface defects in the LIB electrodes. Firstly, based on the change of the gray value in designated area of the electrode surface and the specific ring area obtained according to the defect shape, the data set is divided into bright defects and dark defects. And then Gabor filter is exploited to extract the features of image, and the Random forest(RF) classifier is used for further detailed classification in the two divided categories. Through experimental results, it is found that the rough classification which is the first step in RD-GRF exactly improves the classification accuracy. All in all, in the case of comprehensively considering classification accuracy and speed, RD-GRF achieves the best result compared with the other three famous algorithms.

关键词： Lithium-ion battery defect classification Two-step method Gabor filter Random forest

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

defect IDENTIFICATION BY AN ULTRASONIC CYLINDRICAL PHASED ARRAY

DEFECT IDENTIFICATION BY AN ULTRASONIC CYLINDRICAL PHASED AR...

引用

2016年全国压电和声波理论及器件应用研讨会

作者： Shi-wen CHEN Bi-xing ZHANG State Key Laboratory of Acoustics Institute of Acoustics Chinese Academy of Sciences

ISBN: (纸本)9781509011902

Background, Motivation and Objective Ultrasonic phased array technology is widely used in industrial Non-Destructive Testing(NDT). However, the traditional planar phased array transducer is usually used to scan the object with flat surface. To imaging the inner surface of a borehole wall, an ultrasonic cylindrical phased array transducer is presented to fulfill the whole scanning process for the inner surface. On the other hand, conventional defects identification usually depends on the specific professional knowledge and the experience of the operating personnel which could cause the instability and individual differences of the analysis results easily. In this paper, a new method is presented and studied for defect detection and classification with the cylindrical phased array by the wavelet-packet transform(WPT) and deep neural network. Statement of Contribution/Methods The WPT is used to extract the signal feature of the ultrasonic echo from defects, and the deep neural network is used to classify the defects feature. Firstly, a finite element model is conducted to simulate the defects identification by the cylindrical phased array transducer. A series of simulation are done for two types of defects with different sizes by a 64-element cylindrical phased array transducer with the center frequency of 500 k Hz. In each scanning process, only 8 elements would be excited in the ultrasonic phased array. Forty defects are modeled and studied, twenty of them are cubic and the others are sphere. The cylindrical phased array is used to transmit and receive the echo signals. The ultrasonic echo pulses are reflected from the defects and carried the shape information of the defects. Then, the WPT decompose algorithm is used to four-levers decompose, reconstruct and extract the feature of these echo signals. The reconstructed signals are used to the deep neural network to the defect classification. The output value(1, 0) and(0, 1) represent cubic defect and sphere defects, respectively. The influence of the distance betwe

关键词： Ultrasonic cylindrical phased array defect classification Wavelet-packet transform Deep neural network

来源： cnki会议评论

在线全文

cnki会议

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：