Study on the Laser-Based Weld Surface Flaw Identification System Employing Wavelet Analysis Methodology

摘要

A surface flaw found in weld can potentially pose a safety concern to industrial applications which could result into fatal incidents. To increase the integrity of these structures, a new surface flaw identification technique on enhancing the flaw characterization is studied for weld inspection. The proposed technique employs the integration of a laser-line profile sensor and processing module based on wavelet analysis. The laser-line sensor acquired the two-dimensional profile of a target weld based on the principle of laser triangulation and yield the height and width information of the weld. This two-dimensional profile data is then processed by a symlet wavelet (sym8) based flaw identification algorithm to calculate the positions of the flaws on the sample. Experimental results for different flaw sizes show that the surface flaw identification technique is able to accurately identify and locate surface features 0.2 mm wide.