Real-Time Concrete Crack Detection and Instance Segmentation using Deep Transfer Learning

Lasitha Piyathilaka; D. M. G. Preethichandra; Umer Izhar; Gayan Kahandawa

doi:10.3390/ecsa-7-08260

Back

Real-Time Concrete Crack Detection and Instance Segmentation using Deep Transfer Learning

Conference paper

Open access

Peer reviewed

Real-Time Concrete Crack Detection and Instance Segmentation using Deep Transfer Learning

Lasitha Piyathilaka, D. M. G. Preethichandra, Umer Izhar and Gayan Kahandawa

Engineering Proceedings, Vol.2(1), pp.1-7

International Electronic Conference on Sensors and Applications, 7th (Online, 15-Nov-2020–30-Nov-2020)

MDPI AG

2020

DOI: https://doi.org/10.3390/ecsa-7-08260

Files and links (1)

pdf

engproc-02-000913.66 MBDownload View

Published VersionCC BY V4.0, Open Access

Abstract

crack detection

visual inspection

infrastructure monitoring

deep learning

real-time detection

crack dataset

Cracks on concrete infrastructure are one of the early indications of structural degradation which needs to be identified early as possible to carry out early preventive measures to avoid further damage. In this paper, we propose to use YOLACT: a real-time instance segmentation algorithm for automatic concrete crack detection. This deep learning algorithm is used with transfer learning to train the YOLACT network to identify and localize cracks with their corresponding masks which can be used to identify each crack instance. The transfer learning techniques allowed us to train the network on a relatively small dataset of 500 crack images. To train the YOLACT network, we created a dataset with ground-truth masks from images collected from publicly available datasets. We evaluated the trained YOLACT model for concrete crack detection with ResNet-50 and ResNet-101 backbone architectures for both precision and speed of detection. The trained model achieved high mAP results with real-time frame rates when tested on concrete crack images on a single GPU. The YOLACT algorithm was able to correctly segment multiple cracks with individual instance level masks with high localization accuracy.

Details

Title: Real-Time Concrete Crack Detection and Instance Segmentation using Deep Transfer Learning
Authors: Lasitha Piyathilaka (Corresponding Author) - Central Queensland University
D. M. G. Preethichandra - Central Queensland University
Umer Izhar - University of the Sunshine Coast, Queensland, School of Science, Technology and Engineering
Gayan Kahandawa - Federation University
Publication details: Engineering Proceedings, Vol.2(1), pp.1-7
Conference details: International Electronic Conference on Sensors and Applications, 7th (Online, 15-Nov-2020–30-Nov-2020)
Publisher: MDPI AG
Date published: 2020
DOI: 10.3390/ecsa-7-08260
ISSN: 2673-4591
Copyright note: © 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Organisation Unit: School of Science, Technology and Engineering
Language: English
Record Identifier: 991045198902621
Output Type: Conference paper

Metrics

106 File views/ downloads

56 Record Views