Deep Learning-Based Crack Identification for Steel Pipelines by Extracting Features from 3D Shadow Modeling

Automatic crack identification for pipeline analysis utilizes three-dimensional (3D) image technology to improve the accuracy and reliability of identification. A new technique that integrates a deep learning algorithm 3D shadow modeling (3D-SM) is proposed automatic corrosion cracks in pipelines. Since depth below surrounding area crack, projected when exposed under light sources. In this study, we analyze areas through identify evolving shape shadows. To denoise images, connected domain implemented so groups can be retained scattered occur due insignificant defects eliminated. Moreover, novel neural network developed process images. The method successfully processes images efficiently accurately diagnoses cracks. Experimental results show achieves satisfactory performance with 93.53% 92.04% regression rate.