Segmentation of surface cracks based on a fully convolutional neural network and gated scale pooling

Research output: Chapter in Book/Report/Conference proceedingConference contribution

16 Downloads (Pure)

Abstract

Continual use, as well as aging, allows cracks to develop on concrete surfaces. These cracks are early indications of surface degradation. Therefore, regular inspection of surfaces is an important step in preventive maintenance, allowing reactive measures in a timely manner when cracks may impair the integrity of a structure. Automating parts of this inspection process provides the potential for improved performance and more efficient resource usage, as these inspections are usually carried out manually by trained inspectors. In this work we propose a Fully Convolutional, U-Net based, Neural Network architecture to automatically segment cracks. Conventional pooling operations in Convolutional Neural Networks are static operations that reduce the spatial size of an input, which may lead to loss of information as features are discarded. In this work we introduce and incorporate a novel pooling function into our architecture, Gated Scale Pooling. This operation aims to retain features from multiple scales as well as adapt proactively to the feature map being pooled. Training and testing of our network architecture is conducted on three different public surface crack datasets. It is shown that employing Gated Scale Pooling instead of Max Pooling achieves superior results. Furthermore, our experiments also indicate strongly competitive results when compared with other crack segmentation techniques.
Original languageEnglish
Title of host publication2019 27th European Signal Processing Conference (EUSIPCO)
PublisherIEEE
ISBN (Print)9789082797039
DOIs
Publication statusPublished - 18 Nov 2019

Fingerprint

Cracks
Neural networks
Inspection
Network architecture
Preventive maintenance
Aging of materials
Concretes
Degradation
Testing
Experiments

Keywords

  • crack segmentation
  • deep learning
  • CNN
  • pooling

Cite this

@inproceedings{8449891d4a5c4f649ba2b0240a1baef2,
title = "Segmentation of surface cracks based on a fully convolutional neural network and gated scale pooling",
abstract = "Continual use, as well as aging, allows cracks to develop on concrete surfaces. These cracks are early indications of surface degradation. Therefore, regular inspection of surfaces is an important step in preventive maintenance, allowing reactive measures in a timely manner when cracks may impair the integrity of a structure. Automating parts of this inspection process provides the potential for improved performance and more efficient resource usage, as these inspections are usually carried out manually by trained inspectors. In this work we propose a Fully Convolutional, U-Net based, Neural Network architecture to automatically segment cracks. Conventional pooling operations in Convolutional Neural Networks are static operations that reduce the spatial size of an input, which may lead to loss of information as features are discarded. In this work we introduce and incorporate a novel pooling function into our architecture, Gated Scale Pooling. This operation aims to retain features from multiple scales as well as adapt proactively to the feature map being pooled. Training and testing of our network architecture is conducted on three different public surface crack datasets. It is shown that employing Gated Scale Pooling instead of Max Pooling achieves superior results. Furthermore, our experiments also indicate strongly competitive results when compared with other crack segmentation techniques.",
keywords = "crack segmentation, deep learning, CNN, pooling",
author = "Jacob Konig and Jenkins, {Mark David} and Peter Barrie and Mike Mannion and Gordon Morison",
note = "Acceptance in SAN",
year = "2019",
month = "11",
day = "18",
doi = "10.23919/EUSIPCO.2019.8902341",
language = "English",
isbn = "9789082797039",
booktitle = "2019 27th European Signal Processing Conference (EUSIPCO)",
publisher = "IEEE",

}

Segmentation of surface cracks based on a fully convolutional neural network and gated scale pooling. / Konig, Jacob; Jenkins, Mark David; Barrie, Peter; Mannion, Mike; Morison, Gordon.

2019 27th European Signal Processing Conference (EUSIPCO). IEEE, 2019.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Segmentation of surface cracks based on a fully convolutional neural network and gated scale pooling

AU - Konig, Jacob

AU - Jenkins, Mark David

AU - Barrie, Peter

AU - Mannion, Mike

AU - Morison, Gordon

N1 - Acceptance in SAN

PY - 2019/11/18

Y1 - 2019/11/18

N2 - Continual use, as well as aging, allows cracks to develop on concrete surfaces. These cracks are early indications of surface degradation. Therefore, regular inspection of surfaces is an important step in preventive maintenance, allowing reactive measures in a timely manner when cracks may impair the integrity of a structure. Automating parts of this inspection process provides the potential for improved performance and more efficient resource usage, as these inspections are usually carried out manually by trained inspectors. In this work we propose a Fully Convolutional, U-Net based, Neural Network architecture to automatically segment cracks. Conventional pooling operations in Convolutional Neural Networks are static operations that reduce the spatial size of an input, which may lead to loss of information as features are discarded. In this work we introduce and incorporate a novel pooling function into our architecture, Gated Scale Pooling. This operation aims to retain features from multiple scales as well as adapt proactively to the feature map being pooled. Training and testing of our network architecture is conducted on three different public surface crack datasets. It is shown that employing Gated Scale Pooling instead of Max Pooling achieves superior results. Furthermore, our experiments also indicate strongly competitive results when compared with other crack segmentation techniques.

AB - Continual use, as well as aging, allows cracks to develop on concrete surfaces. These cracks are early indications of surface degradation. Therefore, regular inspection of surfaces is an important step in preventive maintenance, allowing reactive measures in a timely manner when cracks may impair the integrity of a structure. Automating parts of this inspection process provides the potential for improved performance and more efficient resource usage, as these inspections are usually carried out manually by trained inspectors. In this work we propose a Fully Convolutional, U-Net based, Neural Network architecture to automatically segment cracks. Conventional pooling operations in Convolutional Neural Networks are static operations that reduce the spatial size of an input, which may lead to loss of information as features are discarded. In this work we introduce and incorporate a novel pooling function into our architecture, Gated Scale Pooling. This operation aims to retain features from multiple scales as well as adapt proactively to the feature map being pooled. Training and testing of our network architecture is conducted on three different public surface crack datasets. It is shown that employing Gated Scale Pooling instead of Max Pooling achieves superior results. Furthermore, our experiments also indicate strongly competitive results when compared with other crack segmentation techniques.

KW - crack segmentation

KW - deep learning

KW - CNN

KW - pooling

U2 - 10.23919/EUSIPCO.2019.8902341

DO - 10.23919/EUSIPCO.2019.8902341

M3 - Conference contribution

SN - 9789082797039

BT - 2019 27th European Signal Processing Conference (EUSIPCO)

PB - IEEE

ER -