Clifford Pattern Matching for Color Image Edge Detection
2006 (English)In: Visualization of Large and Unstructured Data Sets: first workshop of the DFG's International Research Training Group Visualization of Large and Unstructured Data Sets - Applications in Geospatial Planning, Modeling, and Engineering, June 14-16, 2006, Dagstuhl Castle, Germany / [ed] Hans Hagen, Andreas Kerren, Peter Dannenmann, GI-Edition , 2006, 47-58 p.Chapter in book (Refereed)
Feature detection and pattern matching play an important role in visualization.Originally developed for images and scalar fields, pattern matching methods become increasingly interesting for other applications, e.g., vector fields. To apply pattern matching to vector fields the basic concepts of convolution and fast Fourier transform (FFT) have to be generalized to vector fields. A formalism supporting an elegant generalization of these concepts is provided by the Clifford Algebra, originally developed for describing geometry and geometric operations. We discuss an application of the Clifford Pattern Matching (CPM). We apply CPM to images for ”Clifford Color Edge Detection” (C2ED), an approach for detecting edges and other features in color images. The basic idea is to treat color value tripels as vectors and apply the pattern matching algorithm to the resulting vector field. We introduce vector-valued filters for edge detection and present results.
Place, publisher, year, edition, pages
GI-Edition , 2006. 47-58 p.
Computer Vision and Robotics (Autonomous Systems)
IdentifiersURN: urn:nbn:se:liu:diva-128065ISBN: 978-38-8579-438-7OAI: oai:DiVA.org:liu-128065DiVA: diva2:928799