RT Journal Article
T1 Combining boundary and region features inside the combinatorial pyramid  for topology-preserving perceptual image segmentation
A1 Antúnez, Esther
A1 Marfil-Robles, Rebeca
A1 Bandera-Rubio, Antonio Jesús
K1 Imágenes
K1 Topología
AB Combinatorial pyramids represent the image as a stack of successively reduced combi-natorial maps, which encode the whole image at different levels of abstraction. Withinthis framework, this paper proposes to conduct the perceptual organization of the im-age content in two consecutive stages. The first stage builds the lower set of levels ofthe hierarchy according to simple face (regions) features (colour and size). On the topof this hierarchy, the second stage will mainly employ boundary features, encoded inthe darts of the combinatorial maps, to obtain a second set of levels of abstraction. TheBerkeley data set BSDS300 is used to quantitatively compare the performance of theproposal to a number of perceptual grouping approaches, showing that it yields betteror similar results than most of these algorithms while offering two interesting features:computation at multiple image resolutions and preservation of the image topology.
PB Elsevier
YR 2012
FD 2012
LK https://hdl.handle.net/10630/34515
UL https://hdl.handle.net/10630/34515
LA eng
NO Antúnez, E., Marfil, R., Bandera, A. Combining boundary and region features inside the combinatorial pyramid for topology-preserving perceptual image segmentation. Pattern Recognition Letters, 2012, 33(16), pp. 2245–2253
DS RIUMA. Repositorio Institucional de la Universidad de Málaga
RD 20 ene 2026