Crack Detection and Filling, Using Steepest Descent Method
I.A. Ismail, E. A. Rakh, S.I. Zaki, M. A. Ashabrawy, Mohamed K. Shaat · International Journal of Computer and Electrical Engineering · 2009
An integrated methodology for the detection and removal of cracks on digitized images is presented in this paper. The cracks are detected by steepest descent algorithm (SDA) the output of the piping image as cracks are removed using either a gradient Function (GRF) and processed data or a semi-automatic procedure based on region growing. Finally, crack filling using the steepest descent algorithm. The methodology has been shown to perform very well on digitized images suffering from cracks, using Matlab, Surfer and Visual Fortran programming.