Bridge Crack Detection based on Deep Learning and Drone

Abstract

Crack detection in the bridge is one of the crucial aspects of the evaluation and maintenance of bridges. The existing image-based methods require capturing the surface of the bridge and extracting the crack features to detect the crack. However, in some positions such as the space under the bridge or piers, it is difficult to capture images for crack detection. This paper aims to apply a method to detect cracks on the bridge by using a drone that can capture images in challenging positions. The video recorded from the drone will be automatically identified the cracks by employing the deep learning method. Deep learning is designed for training and testing the dataset with 40.000 images, each image sized 244x244. The deep learning method shows the feasibility of detecting the cracks in the transport facility. This is supported by the high accuracy of the experimental results of 93.6%.