Deep Learning for Segmentation of Cracks in High-Resolution Images of Steel Bridges
This work addresses the need for more effective and cost-efficient bridge inspections for infrastructure maintenance, though it is incremental as it builds on existing methods.
The paper tackles the problem of automating crack detection in steel bridge inspections by developing a deep-learning method that integrates ConvNext with an encoder-decoder network and introduces a loss function to reduce false positives, achieving a significant reduction in false positive rates.
Automating the current bridge visual inspection practices using drones and image processing techniques is a prominent way to make these inspections more effective, robust, and less expensive. In this paper, we investigate the development of a novel deep-learning method for the detection of fatigue cracks in high-resolution images of steel bridges. First, we present a novel and challenging dataset comprising of images of cracks in steel bridges. Secondly, we integrate the ConvNext neural network with a previous state-of-the-art encoder-decoder network for crack segmentation. We study and report, the effects of the use of background patches on the network performance when applied to high-resolution images of cracks in steel bridges. Finally, we introduce a loss function that allows the use of more background patches for the training process, which yields a significant reduction in false positive rates.