Monday, November 13, 2000
11:00 AM
CSB 209
Markus Brandner
Technical U. Graz
Fast, Robust Pattern Matching
The presentation will start with a brief overview of computer vision and pattern recognition activities in Austria, and describe the main areas of the major Austrian research labs. At the Institute of Electrical Measurement and Measurement Signal Processing - EMT - at Graz University of Technology, we focus our research in image analysis on several aspects of real-time vision systems with applications in tracking, object recognition, matching, and navigation. Several current EMT research projects will be introduced briefly, including: real-time tracking for virtual and augmented reality, and active object recognition, as well as applications to fruit grading and medical image analysis. The second part of the talk will present in detail a new, fast and robust matching system for applications in industrial inspection. In this talk a new hierarchical structure for fast, robust geometry-based pattern matching is proposed. As opposed to many pattern matching systems reported in the literature we use a structure comprising a number of alternating feature processing and constraint layers. Feature extraction accuracy ranges from coarse at the bottom of the structure to a fine level at the top. A 2D quasi-affine matching system based on the proposed structure has been implemented. Experiments show the reduction in the amount of image data being processed in every layer of the structure as a consequence of applying constraints to the data between two adjacent feature extraction layers. The structure is able to utilize scalable feature extraction algorithms as well as the incorporation of a priori knowledge into the feature extraction.