Shiny surface scanning and reconstruction in robot-laser systems

• Main Author: Carson, Robyn Christine
• Other Authors: Zhong Wende
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/67108
• Institution: Nanyang Technological University

Laser stripe extraction is used extensively in industrial environments, as it is essential to 3D reconstructions of objects by triangulation laser scanning. However laser stripe extraction is highly sensitive to noise. When laser stripe extraction is used in industrial environments noise and uneven surfaces are very common. These factors will modify the shape of the laser profile and so the reconstructed profile. This project proposes a solution that is a robust and accurate way to reconstruct the profile of a shiny surface when noise is present. Laser stripe extraction is used to reconstruct the profile, firstly using a simple laser peak detection method. A “smoothness” formula is then derived and applied to improve the laser peak detection algorithm.