Title :
Decomposition of two-dimensional shapes by morphological approach
Author_Institution :
Dipt. di Matematica, Cagliari Univ., Italy
fDate :
8/29/2002 12:00:00 AM
Abstract :
A major problem in accessing and retrieving visual information from large databases is to provide efficient matching by combining useful representations and suitable similarity models with efficient perceptual indexing. Here, a morphological approach to decompose a binary shape into tokens corresponding to its protrusions is proposed. Each token can be associated with a set of perceptual features that can be used in indexing image databases
Keywords :
image matching; image representation; image segmentation; image sequences; image thinning; mathematical morphology; visual databases; efficient matching; image databases; morphological approach; morphological shape boundary segmentation; pattern recognition problems; perceptual features; perceptual indexing; similarity models; skeleton partitioning; thinned representation; tokens; two-dimensional shape decomposition; visual information access; visual information retrieval;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20020724
