Publication

Abstract : Aircraft industry relies very much on visual inspection for routine inspection of aircrafts. A skilled inspector goes around the aircraft and checks the surface for damages such as dents, cracks and corrosion. This is however, a laborious task and sometimes maybe ineffective. This raises the need for a system to automate the process in order to make it faster, efficient and reliable. A probe going around the aircraft capturing its images and transferring them to a PC which performs image processing algorithms may be an alternative solution to the current visual inspection method. 2D image processing can also aid in detecting damages, but it also detects surface features such as writings, markings etc as damages which makes it difficult to identify the damage. 3D reconstruction of the surface can give clear cut idea about the surface and it can also help measurement. Fringe projection profilometry is one of the widely used methods to reconstruct surfaces, models, etc where fringe patterns are projected on the surface to be reconstructed and phase information is retrieved from the captured fringe images in order to reconstruct the surface. This necessitates the response of the projector to be linear so that the grayscale variations in the profile are projected properly. However, commercially available projectors are gamma distorted which needs to be corrected first. This paper describes studies carried out to study the effect of gamma correction in the surface reconstruction and the validation of a look up table (LUT) based gamma correction method. As the aircraft surfaces can be of different types, it is required to validate the methodology on different surfaces, with LUT made by capturing projected patterns on different surfaces. The results indicate that the method is independent of the surface used and the projector brightness levels.