Multiphase Local Mean Geodesic Active Regions
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Standard
Multiphase Local Mean Geodesic Active Regions. / Hansen, Jacob Daniel Kirstejn; Lauze, Francois Bernard.
Proceedings, 24th International Conference on Pattern Recognition (ICPR). IEEE, 2018. p. 3031- 3036.Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - Multiphase Local Mean Geodesic Active Regions
AU - Hansen, Jacob Daniel Kirstejn
AU - Lauze, Francois Bernard
PY - 2018/8
Y1 - 2018/8
N2 - This paper presents two variational multiphase segmentation methods for recovery of segments in weakly structured images, presenting local and global intensity bias fields, as often is the case in micro-tomography. The proposed methods assume a fixed number of classes. They use local image averages as discriminative features and binary labelling for class membership and their relaxation to per pixel/voxel posterior probabilities, Hidden Markov Measure Field Models (HMMFM). The first model uses a Total Variation weighted semi-norm (wTV) for label field regularization, similar to Geodesic Active Contours, but with a different and possibly richer representation. The second model uses a weighted Dirichlet (squared gradient) regularization. Both problems are solved by alternating minimization on computation of local class averages and label fields. The quadratic problem is essentially smooth, except for HMMFM constraints. The wTV problem uses a Chambolle-Pock scheme for label field updates. We demonstrate on synthetic examples the capabilities of the approaches, and illustrate it on a real examples.
AB - This paper presents two variational multiphase segmentation methods for recovery of segments in weakly structured images, presenting local and global intensity bias fields, as often is the case in micro-tomography. The proposed methods assume a fixed number of classes. They use local image averages as discriminative features and binary labelling for class membership and their relaxation to per pixel/voxel posterior probabilities, Hidden Markov Measure Field Models (HMMFM). The first model uses a Total Variation weighted semi-norm (wTV) for label field regularization, similar to Geodesic Active Contours, but with a different and possibly richer representation. The second model uses a weighted Dirichlet (squared gradient) regularization. Both problems are solved by alternating minimization on computation of local class averages and label fields. The quadratic problem is essentially smooth, except for HMMFM constraints. The wTV problem uses a Chambolle-Pock scheme for label field updates. We demonstrate on synthetic examples the capabilities of the approaches, and illustrate it on a real examples.
U2 - 10.1109/ICPR.2018.8545684
DO - 10.1109/ICPR.2018.8545684
M3 - Article in proceedings
SP - 3031
EP - 3036
BT - Proceedings, 24th International Conference on Pattern Recognition (ICPR)
PB - IEEE
Y2 - 20 August 2018 through 24 August 2018
ER -
ID: 217393760