Damage Segmentation on High-Resolution Coating Images Using a Novel Two-Stage Network Pipeline

Research output: Contribution to journalArticleResearchpeer review

Authors

View graph of relations

Details

Original languageEnglish
Article number245
JournalAerospace
Volume10
Issue number3
Publication statusPublished - 2 Mar 2023

Abstract

The automation of inspections in aircraft engines is an ever-increasing growing field of research. In particular, the inspection and quantification of coating damages in confined spaces, usually performed manually with handheld endoscopes, comprise tasks that are challenging to automate. In this study, 2D RGB video data provided by commercial instruments are further analyzed in the form of a segmentation of damage areas. For this purpose, large overview images, which are stitched from the video frames, showing the whole coating area are analyzed with convolutional neural networks (CNNs). However, these overview images need to be divided into smaller image patches to keep the CNN architecture at a functional and fixed size, which leads to a significantly reduced field of view (FOV) and therefore a loss of information and reduced network accuracy. A possible solution is a downsampling of the overview image to decrease the number of patches and increase this FOV for each patch. However, while an increased FOV with downsampling or a small FOV without resampling both exhibit a lack of information, these approaches incorporate partly different information and abstractions to be utilized complementary. Based on this hypothesis, we propose a two-stage segmentation pipeline, which processes image patches with different FOV and downsampling factors to increase the overall segmentation accuracy for large images. This includes a novel method to optimize the position of image patches, which leads to a further improvement in accuracy. After a validation of the described hypothesis, an evaluation and comparison of the proposed pipeline and methods against the single-network application is conducted in order to demonstrate the accuracy improvements.

Keywords

    CNN, damage inspection, DeeplabV3+, endoscopic inspection, semantic segmentation, transfer learning

ASJC Scopus subject areas

Cite this

Damage Segmentation on High-Resolution Coating Images Using a Novel Two-Stage Network Pipeline. / Hedrich, Kolja; Hinz, Lennart; Reithmeier, Eduard.
In: Aerospace, Vol. 10, No. 3, 245, 02.03.2023.

Research output: Contribution to journalArticleResearchpeer review

Hedrich K, Hinz L, Reithmeier E. Damage Segmentation on High-Resolution Coating Images Using a Novel Two-Stage Network Pipeline. Aerospace. 2023 Mar 2;10(3):245. doi: 10.3390/aerospace10030245
Download
@article{97e244309416475bb800ec7cbebb0ae4,
title = "Damage Segmentation on High-Resolution Coating Images Using a Novel Two-Stage Network Pipeline",
abstract = "The automation of inspections in aircraft engines is an ever-increasing growing field of research. In particular, the inspection and quantification of coating damages in confined spaces, usually performed manually with handheld endoscopes, comprise tasks that are challenging to automate. In this study, 2D RGB video data provided by commercial instruments are further analyzed in the form of a segmentation of damage areas. For this purpose, large overview images, which are stitched from the video frames, showing the whole coating area are analyzed with convolutional neural networks (CNNs). However, these overview images need to be divided into smaller image patches to keep the CNN architecture at a functional and fixed size, which leads to a significantly reduced field of view (FOV) and therefore a loss of information and reduced network accuracy. A possible solution is a downsampling of the overview image to decrease the number of patches and increase this FOV for each patch. However, while an increased FOV with downsampling or a small FOV without resampling both exhibit a lack of information, these approaches incorporate partly different information and abstractions to be utilized complementary. Based on this hypothesis, we propose a two-stage segmentation pipeline, which processes image patches with different FOV and downsampling factors to increase the overall segmentation accuracy for large images. This includes a novel method to optimize the position of image patches, which leads to a further improvement in accuracy. After a validation of the described hypothesis, an evaluation and comparison of the proposed pipeline and methods against the single-network application is conducted in order to demonstrate the accuracy improvements.",
keywords = "CNN, damage inspection, DeeplabV3+, endoscopic inspection, semantic segmentation, transfer learning",
author = "Kolja Hedrich and Lennart Hinz and Eduard Reithmeier",
note = "Funding Information: The research was funded by the German Federal Ministry of Education and Research as part of the Aviation Research and Technology Program of the Niedersachsen Ministry of Economic Affairs, Employment, Transport and Digitalisation under the grant number ZW 1-80157862. The authors are responsible for the content of this publication.",
year = "2023",
month = mar,
day = "2",
doi = "10.3390/aerospace10030245",
language = "English",
volume = "10",
number = "3",

}

Download

TY - JOUR

T1 - Damage Segmentation on High-Resolution Coating Images Using a Novel Two-Stage Network Pipeline

AU - Hedrich, Kolja

AU - Hinz, Lennart

AU - Reithmeier, Eduard

N1 - Funding Information: The research was funded by the German Federal Ministry of Education and Research as part of the Aviation Research and Technology Program of the Niedersachsen Ministry of Economic Affairs, Employment, Transport and Digitalisation under the grant number ZW 1-80157862. The authors are responsible for the content of this publication.

PY - 2023/3/2

Y1 - 2023/3/2

N2 - The automation of inspections in aircraft engines is an ever-increasing growing field of research. In particular, the inspection and quantification of coating damages in confined spaces, usually performed manually with handheld endoscopes, comprise tasks that are challenging to automate. In this study, 2D RGB video data provided by commercial instruments are further analyzed in the form of a segmentation of damage areas. For this purpose, large overview images, which are stitched from the video frames, showing the whole coating area are analyzed with convolutional neural networks (CNNs). However, these overview images need to be divided into smaller image patches to keep the CNN architecture at a functional and fixed size, which leads to a significantly reduced field of view (FOV) and therefore a loss of information and reduced network accuracy. A possible solution is a downsampling of the overview image to decrease the number of patches and increase this FOV for each patch. However, while an increased FOV with downsampling or a small FOV without resampling both exhibit a lack of information, these approaches incorporate partly different information and abstractions to be utilized complementary. Based on this hypothesis, we propose a two-stage segmentation pipeline, which processes image patches with different FOV and downsampling factors to increase the overall segmentation accuracy for large images. This includes a novel method to optimize the position of image patches, which leads to a further improvement in accuracy. After a validation of the described hypothesis, an evaluation and comparison of the proposed pipeline and methods against the single-network application is conducted in order to demonstrate the accuracy improvements.

AB - The automation of inspections in aircraft engines is an ever-increasing growing field of research. In particular, the inspection and quantification of coating damages in confined spaces, usually performed manually with handheld endoscopes, comprise tasks that are challenging to automate. In this study, 2D RGB video data provided by commercial instruments are further analyzed in the form of a segmentation of damage areas. For this purpose, large overview images, which are stitched from the video frames, showing the whole coating area are analyzed with convolutional neural networks (CNNs). However, these overview images need to be divided into smaller image patches to keep the CNN architecture at a functional and fixed size, which leads to a significantly reduced field of view (FOV) and therefore a loss of information and reduced network accuracy. A possible solution is a downsampling of the overview image to decrease the number of patches and increase this FOV for each patch. However, while an increased FOV with downsampling or a small FOV without resampling both exhibit a lack of information, these approaches incorporate partly different information and abstractions to be utilized complementary. Based on this hypothesis, we propose a two-stage segmentation pipeline, which processes image patches with different FOV and downsampling factors to increase the overall segmentation accuracy for large images. This includes a novel method to optimize the position of image patches, which leads to a further improvement in accuracy. After a validation of the described hypothesis, an evaluation and comparison of the proposed pipeline and methods against the single-network application is conducted in order to demonstrate the accuracy improvements.

KW - CNN

KW - damage inspection

KW - DeeplabV3+

KW - endoscopic inspection

KW - semantic segmentation

KW - transfer learning

UR - http://www.scopus.com/inward/record.url?scp=85151502300&partnerID=8YFLogxK

U2 - 10.3390/aerospace10030245

DO - 10.3390/aerospace10030245

M3 - Article

AN - SCOPUS:85151502300

VL - 10

JO - Aerospace

JF - Aerospace

IS - 3

M1 - 245

ER -

By the same author(s)