Pixel-level crack segmentation of tunnel lining segments based on an encoder-decoder network

Regular detection and repair for lining cracks are necessary to guarantee the safety and stability of tunnels. The development of computer vision has greatly promoted structural health monitoring. This study proposes a novel encoder-decoder structure, CrackRecNet, for semantic segmentation of lining...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Frontiers of Structural and Civil Engineering Jg. 18; H. 5; S. 681 - 698
Hauptverfasser: HOU, Shaokang, OU, Zhigang, HUANG, Yuequn, LIU, Yaoru
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Beijing Higher Education Press 01.05.2024
Springer Nature B.V
Schlagworte:
Boring machines
Cities
Civil Engineering
Coders
Computer vision
Concrete structures
convolutional neural network
Countries
crack detection
Cracks
Drilling & boring machinery
encoder-decoder structure
Engineering
Flaw detection
Image acquisition
Image processing
Image segmentation
Laser range finders
Pixels
Regions
Research Article
Semantic segmentation
Structural health monitoring
Tunnel construction
tunnel lining segment
Tunnel linings
Tunneling shields
encoder–decoder structure
semantic segmentation
tunnel lining segment
crack detection
convolutional neural network
ISSN:2095-2430, 2095-2449
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Regular detection and repair for lining cracks are necessary to guarantee the safety and stability of tunnels. The development of computer vision has greatly promoted structural health monitoring. This study proposes a novel encoder-decoder structure, CrackRecNet, for semantic segmentation of lining segment cracks by integrating improved VGG-19 into the U-Net architecture. An image acquisition equipment is designed based on a camera, 3-dimensional printing (3DP) bracket and two laser rangefinders. A tunnel concrete structure crack (TCSC) image data set, containing images collected from a double-shield tunnel boring machines (TBM) tunnel in China, was established. Through data preprocessing operations, such as brightness adjustment, pixel resolution adjustment, flipping, splitting and annotation, 2880 image samples with pixel resolution of 448 × 448 were prepared. The model was implemented by Pytorch in PyCharm processed with 4 NVIDIA TITAN V GPUs. In the experiments, the proposed CrackRecNet showed better prediction performance than U-Net, TernausNet, and ResU-Net. This paper also discusses GPU parallel acceleration effect and the crack maximum width quantification.
Bibliographie:Document received on :2023-02-09
Document accepted on :2023-07-25
encoder-decoder structure
tunnel lining segment
convolutional neural network
semantic segmentation
crack detection
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:2095-2430
2095-2449
DOI:10.1007/s11709-024-1048-4