Diploma Thesis DIP-2672

BibliographyYu, Kou: Automatie Recognition and Localization of Special Reference Objects in 3D Scan Data.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Diploma Thesis No. 2672 (2008).
50 pages, english.
Abstract

3D laser scanning is gaining applications increasingly in many fields. Scan registration is needed for combining scan data taken from different scanner locations. For this purpose, a global reference system is required. In this paper a newly designed 2D chessboard-like target is introduced, which can be printed on paper with a standard black-and-white printer and used as reference object for the registration. Additionally, each target has a bar-coded area containing its identity. Such a target, even under affine distortion and small rotation, can be automatically recognized using image processing techniques. The correspondence between identical targets in different scans can be determined automatically by a bar code reading algorithm. A set of methods and algorithms are demonstrated for the target recognition and information extraction including a newly developed recursive algorithm for contour follow­ing. Further research is done for the recognition of chessboard targets with arbitrary rotation.

Department(s)University of Stuttgart, Institute of Parallel and Distributed Systems, Image Understanding
Superviser(s)Schanz, PD Dr. Michael; Kramer, Alexander
Entry dateMarch 28, 2024
   Publ. Computer Science