Analysis of sub-pixel laser spot detection in laser triangulation systems

Patrick Kienle,Elif Nallar,Michael H. Köhler,Martin Jakobi,Alexander W. Koch
DOI: https://doi.org/10.1117/12.2525669
2019-06-21
Abstract:Laser spot detection is an important step of laser triangulation and limits its accuracy. Common methods to determine the center position include circle fitting, the Hough transformation, the gray centroid method or Gaussian fitting. As these algorithms were often tested in different set-ups and under various conditions, they generally lead to diverse results. The aim of this contribution is to investigate the algorithms in a more objective and realistic approach. After a short introduction to laser triangulation, basic information about laser spot center detection is given. Fundamental limitations of the laser spot detection are then considered and analyzed. The investigations are followed by evaluations of measurements in a real laser triangulation setup. Through this approach we could show that the influence of the spatial quantization and the quantization due to the limited bit depth of the detector can be in a similar range as the deviation due to speckle noise in the image. By using adapted image processing steps, the performance of the laser spot center determination could be improved significantly.
What problem does this paper attempt to address?