Civil infrastructure crack detection using deep learning and image segmentation-based techniques

Abstract

In this paper, we study the application of deep learning and image segmentation techniques for crack detection in civil infrastructure. We investigate the effectiveness of some deep learning models, such as Inception, ResNet, MobileNet, VGG, and U-Net, in detecting cracks on civil infrastructure surfaces. Additionally, we implement two segmentation models, U-Net and SAM, to enhance the precision of crack extraction from images. Through simulations and comparative analysis, we evaluate the performance of the models in accurately identifying and delineating cracks in civil infrastructure. The results demonstrate the efficacy of the proposed approach in achieving accurate crack detection, which is crucial for ensuring the structural integrity and safety of civil infrastructure. The proposed model achieved an accuracy of 100% and IoU of 0.95.