TY - JOUR
T1 - Fully convolutional networks for automatic pavement crack segmentation
AU - Escalona, Uriel
AU - Arce, Fernando
AU - Zamora, Erik
AU - Sossa, Humberto
N1 - Publisher Copyright:
© 2019 Instituto Politecnico Nacional. All rights reserved.
PY - 2019
Y1 - 2019
N2 - Pavement cracks are an increasing threat to public safety. Automatic pavement crack segmentation remains a very challenging problem due to crack texture inhomogeneity, high outlier potential, large variability of topologies, and so on. Due to this, automatic pavement crack detection has captured the attention of the computer vision community, and a great quantity of algorithms for solving this task have been proposed. In this work, we study a U-Net network and two variants for automatic pavement crack detection. The main contributions of this research are: 1) two U-Net based network variations for automatic pavement crack detection, 2) a series of experiments to demonstrate that the proposed architectures outperform the state-of-the-art for automatic pavement crack detection using two public and well-known challenging datasets: CFD and AigleRN and 3) the code for this approach.
AB - Pavement cracks are an increasing threat to public safety. Automatic pavement crack segmentation remains a very challenging problem due to crack texture inhomogeneity, high outlier potential, large variability of topologies, and so on. Due to this, automatic pavement crack detection has captured the attention of the computer vision community, and a great quantity of algorithms for solving this task have been proposed. In this work, we study a U-Net network and two variants for automatic pavement crack detection. The main contributions of this research are: 1) two U-Net based network variations for automatic pavement crack detection, 2) a series of experiments to demonstrate that the proposed architectures outperform the state-of-the-art for automatic pavement crack detection using two public and well-known challenging datasets: CFD and AigleRN and 3) the code for this approach.
KW - Automatic pavement crack detection
KW - Fully convolutional neural networks
KW - Pavement cracks
UR - http://www.scopus.com/inward/record.url?scp=85069815153&partnerID=8YFLogxK
U2 - 10.13053/CyS-23-2-3047
DO - 10.13053/CyS-23-2-3047
M3 - Artículo
AN - SCOPUS:85069815153
SN - 1405-5546
VL - 23
SP - 451
EP - 460
JO - Computacion y Sistemas
JF - Computacion y Sistemas
IS - 2
ER -