scholarly journals A Partial Hierarchical Model for Online Low-Resolution Wear Particle Images Classification

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Xuxu Guo ◽  
Rui Tan ◽  
Mingyang Yang ◽  
Xinrong He ◽  
Jia Guo ◽  
...  
Keyword(s):  
Neural Network ◽  
Sliding Wear ◽  
Super Resolution ◽  
Wear Particle ◽  
Support Vector ◽  
Wear Particles ◽  
The Third ◽  
Online Recognition ◽  
New Feature ◽  
Particle Images

Wear particle image analysis is an effective method to detect wear condition of mechanical devices. However, the recognition accuracy and recognition efficiency for online wear particle automatic recognition are always mutual restricted because the online wear particle images have almost no texture information and lack clarity. Especially for confusing fatigue wear particles and sliding wear particles, the online recognition is a challenging task. Based on this requirement, a super-resolution reconstruct technique and partial hierarchical convolutional neural network, SR-PHnet, is proposed to classify wear particles in one step. The structure of this network is composed by three modules, one is super-resolution layer module, the second is convolutional neural network classifier module, and the third is support vector machine (SVM) classifier module. The classification result of the second module is partial input to the third module for precision classification of fatigue and sliding particles. In addition, a new feature of radial edge factor (REF) is put forward to target fatigue and sliding wear particles. The test result shows that the new feature has the capability to distinguish fatigue and sliding particles well and time saving. The comparison experiments of the convolution neural network (CNN) method, support vector machine method (SVM) with and without REF feature, and integrated model of back-propagation (BP) and CNN are produced. The comparison results show that the online recognition speed and online recognition rate of the proposed SR-PHnet model in this paper are both improved markedly, especially for fatigue and sliding wear particles.

scholarly journals Morphological residual convolutional neural network (M-RCNN) for intelligent recognition of wear particles from artificial joints

Friction ◽  
2021 ◽  
Author(s):  
Xiaobin Hu ◽  
Jian Song ◽  
Zhenhua Liao ◽  
Yuhong Liu ◽  
Jian Gao ◽  
...  
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Data Augmentation ◽  
Support Vector ◽  
Wear Particles ◽  
Artificial Joints ◽  
Performance Deterioration ◽  
Overall Performance ◽  
Correct Category ◽  
Intelligent Recognition

AbstractFinding the correct category of wear particles is important to understand the tribological behavior. However, manual identification is tedious and time-consuming. We here propose an automatic morphological residual convolutional neural network (M-RCNN), exploiting the residual knowledge and morphological priors between various particle types. We also employ data augmentation to prevent performance deterioration caused by the extremely imbalanced problem of class distribution. Experimental results indicate that our morphological priors are distinguishable and beneficial to largely boosting overall performance. M-RCNN demonstrates a much higher accuracy (0.940) than the deep residual network (0.845) and support vector machine (0.821). This work provides an effective solution for automatically identifying wear particles and can be a powerful tool to further analyze the failure mechanisms of artificial joints.

scholarly journals Ferrography Wear Particles Image Recognition Based on Extreme Learning Machine

2017 ◽  
Vol 2017 ◽  
pp. 1-6 ◽  
Author(s):  
Qiong Li ◽  
Tingting Zhao ◽  
Lingchao Zhang ◽  
Wenhui Sun ◽  
Xi Zhao
Keyword(s):  
Neural Network ◽  
Extreme Learning Machine ◽  
Image Recognition ◽  
Rapid Development ◽  
Texture Features ◽  
Wear Particle ◽  
Training Algorithms ◽  
Wear Particles ◽  
Learning Machine ◽  
Elm Classifier

The morphology of wear particles reflects the complex properties of wear processes involved in particle formation. Typically, the morphology of wear particles is evaluated qualitatively based on microscopy observations. This procedure relies upon the experts’ knowledge and, thus, is not always objective and cheap. With the rapid development of computer image processing technology, neural network based on traditional gradient training algorithm can be used to recognize them. However, the feedforward neural network based on traditional gradient training algorithms for image segmentation creates many issues, such as needing multiple iterations to converge and easy fall into local minimum, which restrict its development heavily. Recently, extreme learning machine (ELM) for single-hidden-layer feedforward neural networks (SLFN) has been attracting attentions for its faster learning speed and better generalization performance than those of traditional gradient-based learning algorithms. In this paper, we propose to employ ELM for ferrography wear particles image recognition. We extract the shape features, color features, and texture features of five typical kinds of wear particles as the input of the ELM classifier and set five types of wear particles as the output of the ELM classifier. Therefore, the novel ferrography wear particle classifier is founded based on ELM.

scholarly journals Research on 3D reconstruction method of wear particle dynamic image based on multi contour space mapping

2021 ◽  
Vol 1207 (1) ◽  
pp. 012017
Author(s):  
Han Wang ◽  
Hongfu Zuo ◽  
Zhenzhen Liu ◽  
Hang Fei ◽  
Yan Liu ◽  
...  
Keyword(s):  
3D Reconstruction ◽  
Dimensional Space ◽  
Wear Particle ◽  
Dynamic Monitoring ◽  
Coordinate Space ◽  
Particle Dynamic ◽  
Reconstruction Method ◽  
Wear Particles ◽  
Monitoring Methods ◽  
Particle Images

Abstract Aiming at the problems of current image monitoring methods of lubrication oil wear particles, this paper designs and builds a dynamic monitoring system for oil wear particles based on microfluidic microscopic images. A contour-based 3D reconstruction method of debris particle images is proposed. The image sequences of rotating wear particles tracked by a single target are used as data, and the contour of the wear particle is extracted and the data is stored. The minimum area external rectangle method is used to correct the rotation of the particle images for the problem of deflection. And an algorithm based on cylindrical coordinate space conversion is used to convert the discrete contour point data into three-dimensional space. Complete the 3D model reconstruction of microfluidic wear particles. The ability to analyse wear particles in oil online monitoring technology is improved, which also shows new ideas for wear status monitoring and fault diagnosis technology.

Third Body Concept and Wear Particle Behavior in Dry Friction Sliding Conditions

2015 ◽  
Vol 640 ◽  
pp. 1-12 ◽  
Author(s):  
Jean Denape
Keyword(s):  
Dry Friction ◽  
Friction And Wear ◽  
Wear Particle ◽  
Phenomenological Approach ◽  
Wear Particles ◽  
Third Body ◽  
Practical Applications ◽  
The Third ◽  
Body Concept ◽  
Synthetic View

The third body concept is a pragmatic tool for analyzing and understanding the friction and wear of sliding materials. This approach is based on the dominating role played by the wear particles under dry sliding conditions. These particles constitute the major part of what is called the third body. The third body concept was introduced by Maurice Godet in the middle of the 70’s and developed by Yves Berthier since the end of the 80’s who added complementary conceptual tools as the tribological triplet, the accommodation mechanisms and the tribological circuit. The aim of this paper is to give a synthetic view of these concepts, which involves mechanical, material and physicochemical subjects. Concrete examples and case studies from various practical applications are given to illustrate the validity and the efficiency of such a phenomenological approach.

scholarly journals A New Transfer Learning Ensemble Model with New Training Methods for Gear Wear Particle Recognition

2022 ◽  
Vol 2022 ◽  
pp. 1-10
Author(s):  
Chunhua Zhao ◽  
zhangwen Lin ◽  
Jinling Tan ◽  
Hengxing Hu ◽  
Qian Li
Keyword(s):  
Transfer Learning ◽  
Wear Particle ◽  
Experimental Results ◽  
The Other ◽  
Particle Identification ◽  
Lubricating Oil ◽  
Support Vector ◽  
Training Methods ◽  
Wear Particles ◽  
Feature Expression

Aiming at solving the acquisition problems of wear particle data of large-modulus gear teeth and few training datasets, an integrated model of LCNNE based on transfer learning is proposed in this paper. Firstly, the wear particles are diagnosed and classified by connecting a new joint loss function and two pretrained models VGG19 and GoogLeNet. Subsequently, the wear particles in gearbox lubricating oil are chosen as the experimental object to make a comparison. Compared with the other four models’ experimental results, the model superiority in wear particle identification and classification is verified. Taking five models as feature extractors and support vector machines as classifiers, the experimental results and comparative analysis reveal that the LCNNE model is better than the other four models because its feature expression ability is stronger than that of the other four models.

Application of numerical descriptors to the characterization of wear particles obtained from joint replacements

1997 ◽  
Vol 211 (1) ◽  
pp. 1-10 ◽  
Author(s):  
G W Stachowiak ◽  
G B Stachowiak ◽  
P Campbell
Keyword(s):  
Image Analysis ◽  
Hip Prosthesis ◽  
Knee Prosthesis ◽  
Particle Morphology ◽  
Wear Particle ◽  
Wear Particles ◽  
Joint Replacements ◽  
Particle Images ◽  
Ultra High Molecular Weight

The application of image analysis techniques to the characterization of wear particles from failed joint replacements has been described. Wear particles were extracted from periprosthetic tissues collected during revision surgery. Chemical digestive methods were used to separate the wear particles from the biological soft material. The particles isolated were examined by optical and scanning electron microscopy. Digitized particle images were analysed on a Macintosh computer by a specially developed software and by the image analysis program ‘Prism’. The following numerical descriptors were used to characterize the particles: particle size, boundary fractal dimension and shape parameters such as form factor, roundness, convexity and aspect ratio. Elemental composition of the particles was determined by energy dispersive X-ray spectroscopy. Three selected types of wear particles were analysed and compared: titanium (Ti)-based and calcium (Ca)-based particles from a hip prosthesis and ultra-high molecular weight polyethylene (UHMWPE) particles from a knee prosthesis. The particles exhibited significantly different sizes and their shape numerical descriptors were also different. From the results obtained it appears that computer image analysis of wear particle morphology can be employed to determine the wear processes occurring in the joints. In some cases, the condition of the joint can also be assessed based on this analysis.

Comparative Study on Wear Particle Colour Classifications Using Various Machine Learning Algorithms

2014 ◽  
Vol 619 ◽  
pp. 347-351 ◽  
Author(s):  
Peerawatt Nunthavarawong
Keyword(s):  
Machine Learning ◽  
Copper Alloys ◽  
Learning Algorithms ◽  
Wear Particle ◽  
Machine Learning Algorithms ◽  
Wear Particles ◽  
Meta Learning ◽  
And Function ◽  
Particle Images ◽  
Worn Surface

Wear particle assessment is one of state-of-the-art in used lubricant analyses. There are three categories, i.e. shape analysis, surface texture analysis, and particle colour analysis. Especially, an analysis of wear particle colour can be induced to identify the material type from the worn surface when surface failures of component occurred in the most commonly used lubricants. The quantification of wear particle colours is essential, which is readily extracted by the image processing. However, the colours of wear particles are often unrecognised by visual examination of the human, although these are probably indicated via RGB colours by personal computer. This article therefore aims to determine the quantitative colour descriptors of wear particles, and to find out the suitable method to classify the particle colour. In present work, the colours of wear particle images were separated with combined HSI and L*a*b* colour models, and were then classified by using machine learning algorithms as a decision-making tool. These tools consist of the Bayesian classifiers, Tree classifiers, Rule classifiers, Lazy classifiers, Meta-learning classifiers, and Function classifiers. By comparing in their tools, the function classifier tool was performed to accurately distinguish the heated metals, steel particles, dark and red oxides, and copper alloys, resulting in more reliable examination than that of the other tools.

Special Issue on Pattern Recognition

2004 ◽  
Vol 8 (2) ◽  
pp. 83-83
Author(s):  
Osamu Hasegawa ◽  
Keyword(s):  
Neural Network ◽  
Pattern Recognition ◽  
Artificial Neural Network ◽  
Support Vector Machines ◽  
Support Vector ◽  
Neural Cells ◽  
Special Issue ◽  
Field Range ◽  
The Third ◽  
Vector Machines

We are pleased to publish this special JACIII issue on pattern recognition featuring 3 invited and 14 regular papers. The first and second concern support vector machines and Bayesian networks by authors who are field experts, and should serve as an introduction to beginners and a resource for researchers. In the third paper, the authors propose an artificial neural network for pattern recognition using "living" neural cells. This paper was invited because the research it deals with is considered an example of the interfield research expected to develop in the 21st century. The remaining 14 regular papers are reviewed and selected from 19 submitted papers. In reviewing and selecting the 14 regular papers, covering a broad field range from basic theory to applied systems, we focused on the originality of each paper and the viewpoints of the authors toward problems and experimental results. This wide-ranging selection should prove both innovative and enlightening to interested readers. We thank Professors Kaoru Hirota and Toshio Fukuda, editors-in-chief of this journal, for their support of this special issue. We also thank the staff of Fuji Technology Press for its invaluable assistance.

Sign in / Sign up

Export Citation Format

Share Document