A Comparative Study on Performance of Deep Learning Models for Vision-based Concrete Crack Detection according to Model Types 


Vol. 34,  No. 6, pp. 50-57, Dec.  2019
10.14346/JKOSOS.2019.34.6.50


PDF
  Abstract

In this study, various types of deep learning models that have been proposed recently are classified according to data input / output types and analyzed to find the deep learning model suitable for constructing a crack detection model. First the deep learning models are classified into image classification model, object segmentation model, object detection model, and instance segmentation model. ResNet-101, DeepLab V2, Faster R-CNN, and Mask R-CNN were selected as representative deep learning model of each type. For the comparison, ResNet-101 was implemented for all the types of deep learning model as a backbone network which serves as a main feature extractor. The four types of deep learning models were trained with 500 crack images taken from real concrete structures and collected from the Internet. The four types of deep learning models showed high accuracy above 94% during the training. Comparative evaluation was conducted using 40 images taken from real concrete structures. The performance of each type of deep learning model was measured using precision and recall. In the experimental result, Mask R-CNN, an instance segmentation deep learning model showed the highest precision and recall on crack detection. Qualitative analysis also shows that Mask R-CNN could detect crack shapes most similarly to the real crack shapes.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

김병현, 김건순, 진수민, 조수진, "A Comparative Study on Performance of Deep Learning Models for Vision-based Concrete Crack Detection according to Model Types," Journal of the Korean Society of Safety, vol. 34, no. 6, pp. 50-57, 2019. DOI: 10.14346/JKOSOS.2019.34.6.50.

[ACM Style]

김병현, 김건순, 진수민, and 조수진. 2019. A Comparative Study on Performance of Deep Learning Models for Vision-based Concrete Crack Detection according to Model Types. Journal of the Korean Society of Safety, 34, 6, (2019), 50-57. DOI: 10.14346/JKOSOS.2019.34.6.50.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

pISSN : 1229 - 9197
eISSN : 1875 - 0052

Impact Factor(2022): 2.5