Norell, Kristin
- Centre for Image Analysis, Swedish University of Agricultural Sciences
Abstract
Two related methods for automatic estimation of the pith position, i.e., the Centre of the annual rings, in wood log end face images are presented. We use images that depict untreated log end faces that are deliberately chosen to include difficulties such as rot, non-circular shape, uncentered pith and dirt. The images are taken with a regular digital camera in sawmill environments. Both presented methods use local orientation and Hough transform to detect the pith position, but two different ways to compute the local orientation are used. The results are promising for both methods. At least one of the methods is fast enough to use on-line at a sawmill. (C) 2008 Elsevier B.V. All rights reserved.
Keywords
image processing; orientation estimation; rotation filter; Quadrature filters; Laplacian pyramid; annual rings
Published in
Computers and Electronics in Agriculture
2008, volume: 63, number: 2, pages: 155-167
Publisher: ELSEVIER SCI LTD
SLU Authors
UKÄ Subject classification
Computer Science
Publication identifier
- DOI: https://doi.org/10.1016/j.compag.2008.02.006
Permanent link to this page (URI)
https://res.slu.se/id/publ/21251