A Combined Strategy Using FMCDM for Textures Segmentation in Hemispherical Images from Forest Environments

摘要

The research undertaken in this work comprises the design of a segmentation strategy to solve the stereoscopic correspondence problem for a specific kind of hemispherical images from forest environments. Images are obtained through an optical system based on fisheye lens. The aim consists on the identification of the textures belonging to tree trunks. This is carried out through a segmentation process which uses the combination of five single classical classifiers using the Multi-Criteria Decision Making method under Fuzzy logic paradigm. The combined proposal formulated in this research work is of unsupervised nature and can be applied to any type of forest environment, with the appropriate adaptations inherent to the segmentation process in accordance with the nature of the forest environment analyzed.