RT Conference Proceedings
T1 3D Segmentation Method for Natural Environments based on a Geometric-Featured Voxel Map
A1 Plaza, Victoria
A1 Ababsa, Fakhr-Eddine
A1 García-Cerezo, Alfonso José
A1 Gómez-Ruiz, José Antonio
K1 Ingeniería de sistemas
AB This work proposes a new segmentation algorithm for three-dimensional dense point clouds and has beenspecially designed for natural environments where the ground is unstructured and may include big slopes, non-flat areas andisolated areas. This technique is based on a Geometric-Featured Voxel map (GFV) where the scene is discretized inconstant size cubes or voxels which are classified in flat surface, linear or tubular structures and scattered or undefinedshapes, usually corresponding to vegetation. Since this is not a point-based technique the computational cost is significantlyreduced, hence it may be compatible with Real-Time applications. The ground is extracted in order to obtain more accurateresults in the posterior segmentation process. The scene is split into objects and a second segmentation in regions insideeach object is performed based on the voxel’s geometric class. The work here evaluates the proposed algorithm in variousversions and several voxel sizes and compares the results with other methods from the literature. For the segmentationevaluation the algorithms are tested on several differently challenging hand-labeled data sets using two metrics, one of whichis novel.
YR 2015
FD 2015-03-26
LK http://hdl.handle.net/10630/9616
UL http://hdl.handle.net/10630/9616
LA eng
NO Universidad de Málaga. Campus de Excelencia Internacional Andalucía Tech.
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 13 abr 2026