Moment-Based Pattern Representation Using Shape and Grayscale Features
IbPRIA '07 Proceedings of the 3rd Iberian conference on Pattern Recognition and Image Analysis, Part I
Metric classifier using multilevel network of templates
Pattern Recognition and Image Analysis
| Hi-index | 0.00 |
In this paper, an original solution for shape representation is proposed which relies on a spatial partitioning approach. The representation selects a discrete set of reference points with respect to which a relationship matrix is computed, accounting for the spatial distribution of shape pixels. This is accomplished at different levels of resolution by a tree based representation. Depending on the number of points, coarse to fine region and boundary shape information are captured. Properties of the representation are discussed and assessed through an experimental evaluation on a set of sample shapes.