Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Cited by Google
Similars in SciELO
Similars in Google
Share
Revista Facultad Nacional de Agronomía Medellín
Print version ISSN 0304-2847
Abstract
BETANCUR ACEVEDO, Julián Andrés; PRIETO ORTIZ, Flavio Augusto and OSORIO LONDONO, Gustavo Adolfo. SEGMENTATION OF COFFEE BEANS BY MEANS OF SEEDED REGION GROWING TECHNIQUES. Rev. Fac. Nac. Agron. Medellín [online]. 2006, vol.59, n.1, pp.3311-3333. ISSN 0304-2847.
Three segmentation systems are presented which use the Seeded Region Growing Technique SRG. The first one, called the Euclidean System, uses a Euclidean distance measure in order to find the region of interest (coffee bean). The ACB-PCB System uses two discontinuity measures called average contrast and peripheral contrast, which are derived from the mean of the color components of the pixels that form the region and those that form two of its boundaries. Following an iterative process, the Average Contrast Boundary ACB and the Peripheral Contrast Boundary PCB are computed for use in performing the coffee bean segmentation. Finally, the Hybrid System uses both information from the principal geometrical components in the scene (provided by a Color Edge Detector) and the average contrast measure. These segmentation tools were applied to coffee images acquired under controlled conditions. Results showed a good performance of the Color Edge Detector, as well as the ACB-PCB and Hybrid systems.
Keywords : Image processing; image segmentation; Seeded Region Growing SRG.