A new mobile convolutional neural network-based approach for pixel-wise road surface crack detection

In recent years, the worldwide road network has grown rapidly and reached an enormous size. Parallel to this size of the road network, the imperfections on the road surfaces are just as large. This has led to a significant increase in road maintenance costs for traffic management departments. Researchers have proposed methods for automatic flaw detection using deep neural networks to reduce road maintenance costs. However, these methods are insufficient in terms of performance and lightness. In this study, a convolutional neural network-based approach is proposed, which provides more successful road surface crack detection and requires less lightweight compared to other studies. Since the performance of the proposed approach is higher and more lightweight, it can be used more effectively and efficiently in mobile systems.