Interpretation of polarization-encoded images using clustering and lab colour space

Authors:
Samia Ainouz;Jihad Zallat;Antonello De Martino
Affiliations:
Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection, UMR, ULP, CNRS, École Nationale Supérieure de Physique de Strasbourg, Université Louis ...;Laboratoire des Sciences de l'Image, de l'Informatique et de la Télédétection, UMR, ULP, CNRS, École Nationale Supérieure de Physique de Strasbourg, Université Louis ...;Laboratoire de Physique des Interfaces et Couches Minces, Ecole Polytechnique, UMR, CNRS, Palaiseau Cedex, France
Venue:
MUSP'06 Proceedings of the 6th WSEAS international conference on Multimedia systems & signal processing
Year:
2006

Citing 3
Cited 0

On the color image segmentation algorithm based on the thresholding and the fuzzy C-means techniques

Pattern Recognition
Unsupervised fuzzy clustering

Pattern Recognition Letters
Digital Image Processing

Digital Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

Physical interpretation of polarization-encoded images is not obvious since the information content is intricately combined in the channels of this imaging modality. Thus, the need for a proper tool that allows the analysis and understanding of polarization-encoded images is of prime interest. In this paper, a novel mapping is proposed between the Poincaré space and a parametric color space where an adequate distance is defined and used in the fuzzy C-means clustering algorithm for classification issues. Moreover, the obtained label map is used as a priori information to provide an ad hoc colour display, which allows a proper interpretation of the resulting colored image at a glance.