Robust Tree-Ring Detection
The study of tree-rings is a common task in dendrology. Usually the rings deliver information about the age of the tree, historic climate conditions and forest densities. Many different techniques exist to perform the tree-ring detection, but they commonl
- PDF / 3,433,319 Bytes
- 11 Pages / 430 x 660 pts Page_size
- 14 Downloads / 211 Views
2
Department of Computer Science, University of Chile, Blanco Encalada 2120, Santiago, Chile mcerda,[email protected] Department of Computer Science, Pontificia Universidad Cat´ olica de Chile, Av. Vicu˜ na Mackenna 4860(143), Santiago, Chile [email protected] 3 INRIA-Loria Laboratory, Campus Scientifique 54506, Vandoeuvre-l`es-Nancy, France
Abstract. The study of tree-rings is a common task in dendrology. Usually the rings deliver information about the age of the tree, historic climate conditions and forest densities. Many different techniques exist to perform the tree-ring detection, but they commonly are semi-automatic. The main idea of this work is to propose an automatic process for the tree-ring detection and compare it with a manual detection made by an expert in dendrology. The proposed technique is based on a variant of the Generalized Hough Transform (GHT) created using a very simple growing model of the tree. The presented automatic algorithm shows tolerance to textured and very noisy images, giving a good tree-ring recognition in most of the cases. In particular, it correctly detects the 80% of the tree-rings in our sample database. Keywords: dendrology, tree-ring, hough transform.
1
Introduction
The tree-rings or annual growth rings are formed in response to seasonal changes. Generally, a tree-ring is composed by two growth zones. In the first part of the growing season, thin-walled cells of large radial diameters are produced (earlywood), while towards the end of the season thick-walled cells of smaller diameter appears (latewood), resulting in a sharp disjunction between growth rings (see Fig. 1). Analysis of tree-rings from cross-sections of the tree (called stem analysis) plays a main role in assessing growth response of trees to environmental factors. Furthermore, stem analysis is used to develop tree growth models to make yield and stand tables, and to reconstruct the entire historical growth record. Hence it has applicability in dendrochronological analysis 1 . The tree-ring analysis is usually made recording the ring-width of four or eight directions on a wood disc, however in some applications it is necessary to record the entire growth ring [1], achieving a better estimation of ring areas. 1
Study of woody plants such as shrubs and lianas.
D. Mery and L. Rueda (Eds.): PSIVT 2007, LNCS 4872, pp. 575–585, 2007. c Springer-Verlag Berlin Heidelberg 2007
576
M. Cerda, N. Hitschfeld-Kahler, and D. Mery
The automatization of the tree-ring recognition process is important because it could make more comparable and reproducible results, currently manually performed by experts. Additionally, an automatic algorithm could reduce the time required to perform the analysis. The automatization of the tree-ring recognition process requires of image analysis, but this is a tough task, because of a wood disc image contains a high level of noise. The noise of the wood disc images comes mainly from the texture and imperfections of the wood, and the acquisition process itself. Another problem is the difficu
Data Loading...
