Abstract Road pavements require periodic maintenance and repair, which as a national infrastructure facility requires an enormous annual. The crack sealing method has been widely used in the implementation of pavement repair and maintenance. Developed countries have recognized the importance of the crack sealing method and have continuously pursued research on the development of automated crack sealing equipment such as ARMM(Automated Road Maintenance Machine), OCCSM(Operator Controlled Crack Sealing Machine), and TTLS(Transfer Tank Longitudinal Sealer) since the early 1990s. In 2004, APCS(Automated Pavement Crack Sealer), which seals routed cracks on the road, was developed in Korea, and since 2009, the development of ACSTM(Automated Crack Sealer with Telescopic Manipulator) for the sealing of non-routed cracks has been underway. Because the non-routed crack is characterized by a very narrow width of 2~3 mm, in comparison to the routed crack, it is necessary to use high-resolution pavement images of over five million pixels in order to detect non-routed cracks. Moreover, it is very effective to employ intelligent algorithms that can distinguish cracks and noise from the high-resolution images using the morphological characteristics of non-routed crack. The purpose of this study is to develop an intelligent algorithm, which can distinguish crack and noise by eliminating the noise, to enable the ACSTM equipment in easy detection of the non-routed cracks. This study subjects the binary high-resolution images of the non-routed cracks to artificial neural network and binary logistic regression analysis for this purpose of intelligently discerning the crack images from the noise. Actual pavement images have been used to compare and verify the accuracy of the proposed algorithm in identifying cracks.
Development of a crack recognition algorithm from non-routed pavement images using artificial neural network and binary logistic regression
KSCE Journal of Civil Engineering ; 20 , 4 ; 1151-1162
2015-06-22
12 pages
Article (Journal)
Electronic Resource
English
Predictive Models for Prostate Cancer based on Logistic Regression and Artificial Neural Network
British Library Conference Proceedings | 2015
|Human Activity Recognition Using Multinomial Logistic Regression
British Library Conference Proceedings | 2015
|Recognition of asphalt pavement crack length using deep convolutional neural networks
British Library Online Contents | 2018
|