TY - JOUR AU - Herrera, Rodrigo Javier AU - Roncancio, Henry A AU - Velasco Peña, Hugo Fernando AU - Polanco Tapia, César Augusto AU - Rosado, Jenny PY - 2001/11/30 Y2 - 2024/03/28 TI - Segmentación de imágenes de microanatomía de madera mediante una clasificación multiescala basada en operaciones morfológicas JF - Ingeniería JA - Ing. VL - 7 IS - 2 SE - Science, research, academia and development DO - 10.14483/23448393.2820 UR - https://revistas.udistrital.edu.co/index.php/reving/article/view/2820 SP - 64-72 AB - In this work we seek to study wood in more detail to understand better its nature, due to is an organic material with a wide range of applications. There are some methods and tools used in the wood technology laboratories to study properties and potential uses of the species. In this paper we propose a method based on digital image processing, supported on a computational tool, for improving the analysis of these characteristics. In this initial proposal, we show a segmentation of wood microanatomy, directed to identify microanatomies structures for quantifying its concentration into the sample. The used method consists on the creation of a scale space based on morphologic operations and a technique of clustering using the fuzzy c-means algorithm. The multiscale space is created by applying open-close and close-open morphologic operations over images. The morphologic operations are based on a structuring element, which change its size in order to vary scale. Scales are determined by image granulometry study using FFT. We show how the method can be useful for wood microanatomy textures segmentation. ER -