Encoder–decoder network for pixel-level road crack detection in black-box images

Seongdeok Bang, Somin Park, Hongjo Kim, Hyoungkwan Kim

Research output: Contribution to journalArticle

Abstract

Timely monitoring of pavement cracks is essential for successful maintenance of road infrastructure. Accurate information concerning crack location and severity enables proactive management of the infrastructure. Black-box cameras, which are becoming increasingly widespread at an affordable price, can be used as efficient road-image collectors over a wide area. However, the cracks in these images are difficult to detect, because the images containing them often include objects other than roads. Thus, we propose a pixel-level detection method for identifying road cracks in black-box images using a deep convolutional encoder–decoder network. The encoder consists of convolutional layers of the residual network for extracting crack features, and the decoder consists of deconvolutional layers for localizing the cracks in an input image. The proposed network was trained on 427 out of 527 images extracted from black-box videos and tested on the remaining 100 images. Compared with VGG-16, ResNet-50, ResNet-101, ResNet-200 with transfer learning, and ResNet-152 without transfer learning, ResNet-152 with transfer learning exhibited the best performance, achieving recall, precision, and intersection of union of 71.98%, 77.68%, and 59.65%, respectively. The experimental results prove that the proposed method is optimal for detecting cracks in black-box images at the pixel level.

Original languageEnglish
Pages (from-to)713-727
Number of pages15
JournalComputer-Aided Civil and Infrastructure Engineering
Volume34
Issue number8
DOIs
Publication statusPublished - 2019 Aug 1

Fingerprint

Crack detection
Pixels
Cracks
Pavements
Cameras
Monitoring

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Computational Theory and Mathematics

Cite this

@article{dc6bbef6bbac42a69fe7d2ef009cca80,
title = "Encoder–decoder network for pixel-level road crack detection in black-box images",
abstract = "Timely monitoring of pavement cracks is essential for successful maintenance of road infrastructure. Accurate information concerning crack location and severity enables proactive management of the infrastructure. Black-box cameras, which are becoming increasingly widespread at an affordable price, can be used as efficient road-image collectors over a wide area. However, the cracks in these images are difficult to detect, because the images containing them often include objects other than roads. Thus, we propose a pixel-level detection method for identifying road cracks in black-box images using a deep convolutional encoder–decoder network. The encoder consists of convolutional layers of the residual network for extracting crack features, and the decoder consists of deconvolutional layers for localizing the cracks in an input image. The proposed network was trained on 427 out of 527 images extracted from black-box videos and tested on the remaining 100 images. Compared with VGG-16, ResNet-50, ResNet-101, ResNet-200 with transfer learning, and ResNet-152 without transfer learning, ResNet-152 with transfer learning exhibited the best performance, achieving recall, precision, and intersection of union of 71.98{\%}, 77.68{\%}, and 59.65{\%}, respectively. The experimental results prove that the proposed method is optimal for detecting cracks in black-box images at the pixel level.",
author = "Seongdeok Bang and Somin Park and Hongjo Kim and Hyoungkwan Kim",
year = "2019",
month = "8",
day = "1",
doi = "10.1111/mice.12440",
language = "English",
volume = "34",
pages = "713--727",
journal = "Computer-Aided Civil and Infrastructure Engineering",
issn = "1093-9687",
publisher = "Wiley-Blackwell",
number = "8",

}

Encoder–decoder network for pixel-level road crack detection in black-box images. / Bang, Seongdeok; Park, Somin; Kim, Hongjo; Kim, Hyoungkwan.

In: Computer-Aided Civil and Infrastructure Engineering, Vol. 34, No. 8, 01.08.2019, p. 713-727.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Encoder–decoder network for pixel-level road crack detection in black-box images

AU - Bang, Seongdeok

AU - Park, Somin

AU - Kim, Hongjo

AU - Kim, Hyoungkwan

PY - 2019/8/1

Y1 - 2019/8/1

N2 - Timely monitoring of pavement cracks is essential for successful maintenance of road infrastructure. Accurate information concerning crack location and severity enables proactive management of the infrastructure. Black-box cameras, which are becoming increasingly widespread at an affordable price, can be used as efficient road-image collectors over a wide area. However, the cracks in these images are difficult to detect, because the images containing them often include objects other than roads. Thus, we propose a pixel-level detection method for identifying road cracks in black-box images using a deep convolutional encoder–decoder network. The encoder consists of convolutional layers of the residual network for extracting crack features, and the decoder consists of deconvolutional layers for localizing the cracks in an input image. The proposed network was trained on 427 out of 527 images extracted from black-box videos and tested on the remaining 100 images. Compared with VGG-16, ResNet-50, ResNet-101, ResNet-200 with transfer learning, and ResNet-152 without transfer learning, ResNet-152 with transfer learning exhibited the best performance, achieving recall, precision, and intersection of union of 71.98%, 77.68%, and 59.65%, respectively. The experimental results prove that the proposed method is optimal for detecting cracks in black-box images at the pixel level.

AB - Timely monitoring of pavement cracks is essential for successful maintenance of road infrastructure. Accurate information concerning crack location and severity enables proactive management of the infrastructure. Black-box cameras, which are becoming increasingly widespread at an affordable price, can be used as efficient road-image collectors over a wide area. However, the cracks in these images are difficult to detect, because the images containing them often include objects other than roads. Thus, we propose a pixel-level detection method for identifying road cracks in black-box images using a deep convolutional encoder–decoder network. The encoder consists of convolutional layers of the residual network for extracting crack features, and the decoder consists of deconvolutional layers for localizing the cracks in an input image. The proposed network was trained on 427 out of 527 images extracted from black-box videos and tested on the remaining 100 images. Compared with VGG-16, ResNet-50, ResNet-101, ResNet-200 with transfer learning, and ResNet-152 without transfer learning, ResNet-152 with transfer learning exhibited the best performance, achieving recall, precision, and intersection of union of 71.98%, 77.68%, and 59.65%, respectively. The experimental results prove that the proposed method is optimal for detecting cracks in black-box images at the pixel level.

UR - http://www.scopus.com/inward/record.url?scp=85062338764&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062338764&partnerID=8YFLogxK

U2 - 10.1111/mice.12440

DO - 10.1111/mice.12440

M3 - Article

VL - 34

SP - 713

EP - 727

JO - Computer-Aided Civil and Infrastructure Engineering

JF - Computer-Aided Civil and Infrastructure Engineering

SN - 1093-9687

IS - 8

ER -