Share Email Print

Proceedings Paper

Spot detection using an adaptive pyramid algorithm: application to apple’s maturity state
Author(s): Eric Laemmer; Aline Deruyver; Malgorzata Sowinska
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

In this article, an image processing method to detect automatically maturity spots on apples in order to assess their maturity state, is presented. Two different techniques of segmentation are studied. The first one is mono dimensional and looks for the maxima in a sequence of pixels. The other one is a morphological technique called the watershed algorithm. The advantages and the drawbacks of the two techniques are compared. These both approaches are particularly adapted to spot detection but they are sensitive to noise and often provide an over-segmentation. To overcome this difficulty we propose a multi-scale analysis based on adjacency graph pyramid. This method has the advantage of providing a unique segmentation, which does not depend on the order of the regions scanning. The merging will be driven by criteria of a model describing spot characteristics. Experimental results have been obtained on a set of 1200 images of apples, showing the interest of the method.

Paper Details

Date Published: 1 August 2003
PDF: 6 pages
Proc. SPIE 4948, 25th International Congress on High-Speed Photography and Photonics, (1 August 2003); doi: 10.1117/12.516728
Show Author Affiliations
Eric Laemmer, Lab. PHASE, CNRS (France)
Aline Deruyver, Lab. PHASE, CNRS (France)
Malgorzata Sowinska, Lab. PHASE, CNRS (France)

Published in SPIE Proceedings Vol. 4948:
25th International Congress on High-Speed Photography and Photonics
Claude Cavailler; Graham P. Haddleton; Manfred Hugenschmidt, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?