Patch-Based Crack Detection in Black Box Images Using Convolutional Neural Networks

Somin Park, Seongdeok Bang, Hongjo Kam, Hyoungkwan Kim

Research output: Contribution to journalArticle

Abstract

Cracks cause deterioration of road performance and functional or structural failure if not managed in a timely manner. This paper proposes an automated crack detection method using a car black box camera to address this problem. The proposed method uses a deep learning model [i.e., convolutional neural network (CNN)] composed of segmentation and classification modules. The segmentation process is performed to extract only the road surface in order to remove elements that interfere with crack detection in the black box image. Then, cracks are detected through analysis of patch units within the extracted road surface. The proposed CNN architecture classifies the elements of the road surface into three categories (i.e., crack, road marking, and intact area) with 90.45% accuracy. The results of the proposed CNN architecture are better than those of previous studies.

Original languageEnglish
Article number04019017
JournalJournal of Computing in Civil Engineering
Volume33
Issue number3
DOIs
Publication statusPublished - 2019 May 1

Fingerprint

Crack detection
Cracks
Neural networks
Network architecture
Deterioration
Railroad cars
Cameras

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Computer Science Applications

Cite this

@article{a7e787b763b1407dad99fa41ecd98b13,
title = "Patch-Based Crack Detection in Black Box Images Using Convolutional Neural Networks",
abstract = "Cracks cause deterioration of road performance and functional or structural failure if not managed in a timely manner. This paper proposes an automated crack detection method using a car black box camera to address this problem. The proposed method uses a deep learning model [i.e., convolutional neural network (CNN)] composed of segmentation and classification modules. The segmentation process is performed to extract only the road surface in order to remove elements that interfere with crack detection in the black box image. Then, cracks are detected through analysis of patch units within the extracted road surface. The proposed CNN architecture classifies the elements of the road surface into three categories (i.e., crack, road marking, and intact area) with 90.45{\%} accuracy. The results of the proposed CNN architecture are better than those of previous studies.",
author = "Somin Park and Seongdeok Bang and Hongjo Kam and Hyoungkwan Kim",
year = "2019",
month = "5",
day = "1",
doi = "10.1061/(ASCE)CP.1943-5487.0000831",
language = "English",
volume = "33",
journal = "Journal of Computing in Civil Engineering",
issn = "0887-3801",
publisher = "American Society of Civil Engineers (ASCE)",
number = "3",

}

Patch-Based Crack Detection in Black Box Images Using Convolutional Neural Networks. / Park, Somin; Bang, Seongdeok; Kam, Hongjo; Kim, Hyoungkwan.

In: Journal of Computing in Civil Engineering, Vol. 33, No. 3, 04019017, 01.05.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Patch-Based Crack Detection in Black Box Images Using Convolutional Neural Networks

AU - Park, Somin

AU - Bang, Seongdeok

AU - Kam, Hongjo

AU - Kim, Hyoungkwan

PY - 2019/5/1

Y1 - 2019/5/1

N2 - Cracks cause deterioration of road performance and functional or structural failure if not managed in a timely manner. This paper proposes an automated crack detection method using a car black box camera to address this problem. The proposed method uses a deep learning model [i.e., convolutional neural network (CNN)] composed of segmentation and classification modules. The segmentation process is performed to extract only the road surface in order to remove elements that interfere with crack detection in the black box image. Then, cracks are detected through analysis of patch units within the extracted road surface. The proposed CNN architecture classifies the elements of the road surface into three categories (i.e., crack, road marking, and intact area) with 90.45% accuracy. The results of the proposed CNN architecture are better than those of previous studies.

AB - Cracks cause deterioration of road performance and functional or structural failure if not managed in a timely manner. This paper proposes an automated crack detection method using a car black box camera to address this problem. The proposed method uses a deep learning model [i.e., convolutional neural network (CNN)] composed of segmentation and classification modules. The segmentation process is performed to extract only the road surface in order to remove elements that interfere with crack detection in the black box image. Then, cracks are detected through analysis of patch units within the extracted road surface. The proposed CNN architecture classifies the elements of the road surface into three categories (i.e., crack, road marking, and intact area) with 90.45% accuracy. The results of the proposed CNN architecture are better than those of previous studies.

UR - http://www.scopus.com/inward/record.url?scp=85062373204&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062373204&partnerID=8YFLogxK

U2 - 10.1061/(ASCE)CP.1943-5487.0000831

DO - 10.1061/(ASCE)CP.1943-5487.0000831

M3 - Article

AN - SCOPUS:85062373204

VL - 33

JO - Journal of Computing in Civil Engineering

JF - Journal of Computing in Civil Engineering

SN - 0887-3801

IS - 3

M1 - 04019017

ER -