Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods

The accurate characterization of wear debris is crucial for assessing the health of rotating engine components and for conducting simulation experiments in debris detection. This study proposed an intelligent recognition method for ferrography wear debris images, leveraging several improved Mask Reg...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiangwen Xiao, Weixuan Zhang, Qing Wang, Yuan Liu, Yishou Wang
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Lubricants
Subjects:
Online Access:https://www.mdpi.com/2075-4442/13/5/208
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849327195946745856
author Xiangwen Xiao
Weixuan Zhang
Qing Wang
Yuan Liu
Yishou Wang
author_facet Xiangwen Xiao
Weixuan Zhang
Qing Wang
Yuan Liu
Yishou Wang
author_sort Xiangwen Xiao
collection DOAJ
description The accurate characterization of wear debris is crucial for assessing the health of rotating engine components and for conducting simulation experiments in debris detection. This study proposed an intelligent recognition method for ferrography wear debris images, leveraging several improved Mask Region-based Convolutional Neural Network (Mask R-CNN) algorithms to quantitatively calculate both the number of debris particles and their coverage areas. The improvement on the Mask R-CNN focuses on two key aspects: enhancing feature extraction through the feature pyramid network structure and integrating attention mechanisms. The most suitable attention mechanism for wear debris detection was determined through ablation experiments. The improved Mask R-CNN combined with the Convolutional Block Attention Module achieves the best Mean Pixel Accuracy of 87.63% at a processing speed of 7.6 frames per second, demonstrating its high accuracy and efficiency in wear particle segmentation. Furthermore, the quantitative and qualitative analysis of wear debris, including the number and area of debris particles and their classification, provides valuable insights into the severity of wear. These insights are essential for understanding the extent of wear damage and guiding maintenance decisions.
format Article
id doaj-art-e12dbac556954261bfa8d15e4b605229
institution Kabale University
issn 2075-4442
language English
publishDate 2025-05-01
publisher MDPI AG
record_format Article
series Lubricants
spelling doaj-art-e12dbac556954261bfa8d15e4b6052292025-08-20T03:47:57ZengMDPI AGLubricants2075-44422025-05-0113520810.3390/lubricants13050208Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN MethodsXiangwen Xiao0Weixuan Zhang1Qing Wang2Yuan Liu3Yishou Wang4School of Aerospace Engineering, Xiamen University, Xiamen 361005, ChinaSchool of Aerospace Engineering, Xiamen University, Xiamen 361005, ChinaXiamen Airlines, Xiamen 361003, ChinaAECC Hunan Aviation Powerplant Research Institute, Zhuzhou 412000, ChinaSchool of Aerospace Engineering, Xiamen University, Xiamen 361005, ChinaThe accurate characterization of wear debris is crucial for assessing the health of rotating engine components and for conducting simulation experiments in debris detection. This study proposed an intelligent recognition method for ferrography wear debris images, leveraging several improved Mask Region-based Convolutional Neural Network (Mask R-CNN) algorithms to quantitatively calculate both the number of debris particles and their coverage areas. The improvement on the Mask R-CNN focuses on two key aspects: enhancing feature extraction through the feature pyramid network structure and integrating attention mechanisms. The most suitable attention mechanism for wear debris detection was determined through ablation experiments. The improved Mask R-CNN combined with the Convolutional Block Attention Module achieves the best Mean Pixel Accuracy of 87.63% at a processing speed of 7.6 frames per second, demonstrating its high accuracy and efficiency in wear particle segmentation. Furthermore, the quantitative and qualitative analysis of wear debris, including the number and area of debris particles and their classification, provides valuable insights into the severity of wear. These insights are essential for understanding the extent of wear damage and guiding maintenance decisions.https://www.mdpi.com/2075-4442/13/5/208instance segmentationcomputer visionwear debris analysisferrographyconvolutional neural networkmulti-feature fusion
spellingShingle Xiangwen Xiao
Weixuan Zhang
Qing Wang
Yuan Liu
Yishou Wang
Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods
Lubricants
instance segmentation
computer vision
wear debris analysis
ferrography
convolutional neural network
multi-feature fusion
title Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods
title_full Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods
title_fullStr Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods
title_full_unstemmed Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods
title_short Intelligent Recognition Method for Ferrography Wear Debris Images Using Improved Mask R-CNN Methods
title_sort intelligent recognition method for ferrography wear debris images using improved mask r cnn methods
topic instance segmentation
computer vision
wear debris analysis
ferrography
convolutional neural network
multi-feature fusion
url https://www.mdpi.com/2075-4442/13/5/208
work_keys_str_mv AT xiangwenxiao intelligentrecognitionmethodforferrographyweardebrisimagesusingimprovedmaskrcnnmethods
AT weixuanzhang intelligentrecognitionmethodforferrographyweardebrisimagesusingimprovedmaskrcnnmethods
AT qingwang intelligentrecognitionmethodforferrographyweardebrisimagesusingimprovedmaskrcnnmethods
AT yuanliu intelligentrecognitionmethodforferrographyweardebrisimagesusingimprovedmaskrcnnmethods
AT yishouwang intelligentrecognitionmethodforferrographyweardebrisimagesusingimprovedmaskrcnnmethods