Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Similars in SciELO
Share
Computación y Sistemas
On-line version ISSN 2007-9737Print version ISSN 1405-5546
Abstract
LIZARRAGA MORALES, Rocío; SANCHEZ YANEZ, Raúl E. and AYALA RAMIREZ, Víctor. Periodicity and Texel Size Estimation of Visual Texture Using Entropy Cues. Comp. y Sist. [online]. 2011, vol.14, n.3, pp.303-319. ISSN 2007-9737.
Texture periodicity and texture element (texel) size are important characteristics for texture recognition and discrimination. In this paper, an approach to determine both, texture periodicity and texel size, is proposed. Our method is based on the entropy, a texture measure computed from the Sum and Difference Histograms. The entropy value is sensitive to the parameters in such histograms and takes its lowest value when the parameters match with texel size or its integer multiples, in any specific direction. We show the performance of our method by texture synthesis, tiling a sample of the detected size and measuring the similarity between the original image and the synthesized one, showing good results with regular textures and texels with different shapes, being useful for practical applications as well because of its simple implementation.
Keywords : Image Processing; Texture; Size and Shape; Texture Periodicity; Entropy.