Author : Mohamed Abolella Abdellatif Gaber
CoAuthors : Yasser S. Mohameda, Hesham M. Shehataa,Taher H. Awad
Source : Alexandria University Engineering journal by Elsevier
Date of Publication : 10/2019
Abstract :
Automatic crack detection is needed to reduce cost and to improve quality of surfaceinspection that is needed for maintenance of infrastructures. In this research, a novel system wasdeveloped to detect steel cracks and to estimate their depth from 2D images. The objective is todevelop an affordable and user-friendly inspection system in replacement of expensive 3D measure-ment devices. A learning strategy was adopted and several learning structures were exploited todecide on the suitable structure. The average intensities of 2D steel crack profiles was fed to neuralnetwork together with the maximum depth of steel cracks measured by laser microscope to train alearning structure. Feed forward back propagation Neural Network was found to produce an over-all average error of 18.81% in testing which is 10% less than the previous error using another learn-ing strategy (updated 3D Make toolbox) for depth recovery. The system performance is comparableto the state of the art and provides an applicable and affordable inspection device.Ó2019 Faculty of Engineering, Alexandria University. Production and hosting by Elsevier B.V. Thisis anopen access article under the CC BY-NC-ND license
Download PDF