[Back]

G. Biegelbauer, M. Vincze:
"Fast and Robust Bore Detection of Range Image Data";
Talk: Workshop of the Austrian Association for Pattern Recognition, Hagenberg; 2004-06-17 - 2004-06-18; in: "Digital Imaging in Media and Education", (2004), ISBN: 3-85403-179-3; 221 - 228.

This paper presents a fast and robust method to precisely segment and locate bore holes of 4
to 50mm diameter. The task is solved by a robot moving a compact triangulation scanning
sensor to all sides of the object and scanning the bores. Exploiting task the knowledge about
the expected bore diameter and bore pose makes it possible to develop highly robust algorithms
for an industrial application. Sparse data of the bore hole is su±cient to segment the bore
independent of bore chamfer type using a robust normal vector ¯t and a classi¯cation based
on the Gaussian image. A sequential algorithm to ¯t the bore cylinder makes it possible to
calculate the bore pose in less than 1 second. Experiments demonstrate that 120 degrees of
the bore surface are su±cient for robust localization within 0.3mm and 0.5 degrees even in the
presence of ghost points and notches in the bores.

http://aleph.ub.tuwien.ac.at/F?base=tuw01&func=find-c&ccl_term=AC04967692