Segmentation d'images couleur par coalescence non supervisée d'histogrammes 2D et fusion de régions selon la théorie de Dempster-Shafer

Segmentation d'images couleur par coalescence non supervisée d'histogrammes 2D et fusion de régions selon la théorie de Dempster-Shafer

Color image segmentation by unsupervised 2D histogram clustering and Dempster-Shafer region merging

Olivier Lezoray Christophe Charrier 

LUSAC EA 2607, Équipe Vision et Analyse d'Image, IUT SRC, 120 rue de l'exode, 50000 Saint-Lô (France)

Corresponding Author Email: 
o.lezoray@chbg.unicaen.fr
Page: 
605-621
|
Received: 
15 June 2004
|
Accepted: 
N/A
|
Published: 
31 October 2004
| Citation

OPEN ACCESS

Abstract: 

In this paper, a color image segmentation method based on a new approach called bimarginal is proposed.To overcome the drawbacks of the classical marginal approaches, color components are considered in pairs in order to have a partial view of their inner correlation. Working with color images, the three possible combinations are considered as three independant information sources. Each pairwise component combination is firstly analyzed according to an unsupervised morphologic clustering which looks for the dominant colors of a 2D histogram. This leads to obtain three segmentation maps combined by intersection after being simplified. The intersection process itself producing an over-segmentation of the image, a pairwise region merging is done according to a similarity criterion with the Dempster-Shafer theory up to a termination criterion. To fully automate the segmentation, an energy function is proposed to quantify the segmentation quality. The latter acts as a performance indicator and is used all over the segmentation to tune its parameters.

Résumé

Dans cet article nous proposons une méthode de segmentation d'images couleur selon une nouvelle approche que nous appelons bi-marginale. Afin de pallier les défauts des approches marginales classiques, nous considérons les composantes couleur deux à deux afin d'avoir une vue partielle de leur corrélation. Travaillant selon cette vision bi-composante, nous considérons les trois combinaisons possible comme trois sources d'informations indépendantes. Chaque information bi-composante est tout d'abord analysée selon un schéma de coalescence morphologique non supervisé qui recherche les couleurs dominantes d'un histogramme bidimensionnel. Cela permet de construire trois cartes de segmentation distinctes qui sont combinées par intersection après avoir été simplifiées. L'intersection produisant une sur-segmentation, une fusion des régions deux à deux est opérée selon un critère de similarité et selon la combinaison de Dempster-Shafer jusqu'à un critère de terminaison. Afin d'automatiser la méthode de segmentation, une mesure d'énergie est proposée afin de quantifier la qualité d'une segmentation, celle-ci sert tout au long de la méthode proposée comme indicateur de performance de la segmentation afin d'en régler les différents paramètres.

Keywords: 

Color, segmentation, clustering, fusion, Dempster-Shafer, multi-scale, quality

Mots clés

Couleur, segmentation, coalescence, fusion, Dempster-Shafer, multi-échelle, qualité

1. Introduction
2. Coalescence D'histogrammes
3. Fusion De Cartes De Segmentation Selon La Théorie De Dempster-Shafer
4. Conclusion
  References

[App91] A. APPRIOU, Probabilités et incertitude en fusion de données multisenseurs. Revue scientifique et technique de la défense, 11:27-40, 1991.

[BCS98] M. BORSOTTI, P. CAMPADELLI, and R. SCHETTINI, Quantitative evaluation of color image segmentation results. Pattern recognition letters, 19:741-747, 1998.

[BLC03] N. BOULOUDANI, P. LAMBERT, and D. COQUIN, Segmentation automatique des images couleur à base d'indicateurs de performance. In CORESA'03, 2003.

[BVZ01] Y. BOYKOV, O. VEKSLER, and R. ZABIH, Fast approximate energy minimization via graph cuts. IEEE transactions on Pattern Analysis and Machine Intelligence, 23(11):1222-1239, Novembre 2001.

[CCFM04] A.-S. CAPELLE, O. COLOT, and C. FERNANDEZMALOIGNE, Evidential segmentation scheme of multi-echo mr images for the detection of brain tumors using neighborhood information, Information Fusion, 5(3) :203-216, 2004.

[Cel90] M. CELENK, A color clustering technique for image segmentation. Computer Vision Graphics and Image Processing, 52:145-170, 1990.

[CJ04] C. CHARRIER and A-L. JOUSSELME, Color space combination and approximation for vector quantization. In ISIVC'04, pages 231-234, Brest, September 2004.

[Dem67] A. DEMPSTER, Upper and Lower Probablilities Induced by Multivalued Mapping. Ann. Math. Statist., 38:325-339, 1967.

[Den95] T. DENOUEX, A k-nearest neighbor classification rule based on dempster-shafer theory. IEEE Trans. on Systems, Man and Cybernetics, 25(5):804-813, 1995.

[Fon04] M. FONTAINE, Segmentation non supervisée d’images par analyse de la connexité des pixels, PhD thesis, Université des Sciences et Techniques de Lille, 2004.

[GBLP00] A. GILLET, L. MACAIRE, C. BOTTE-LECOCQ, and J-G. POSTAIRE, Fuzzy unsupervised color image segmentation. In Proceedings of CGIP, pages 141-146, 2000.

[GMC01] L. GUIGUES, H. LE MEN, and J-P. COCQUEREZ, Segmentation d'images par minimisation d'un critère mdl dans une pyramide de segmentations. In Proceedings of GRETSI'2001, 2001.

[GMC03] L. GUIGUES, H. LE MEN, and J.-P. COCQUEREZ, Analyse et représentation ensembles-échelle d’une image, in Proceedings of GRETSI’2003, 2003.

[GSD01a] T. GÉRAUD, P.Y. STRUB, and J. DARBON, Color image segmentation based on automatic morphological clustering. In IEEE International Conference on Image Processing, volume 3, pages 70-73, 2001.

[GSD01b] T. GÉRAUD, P.Y. STRUB, and J. DARBON, Segmentation d'images en couleur par classification morphologique non supervisée. In International Conference on Image and Signal Processing (ICISP), pages 387-394, 2001.

[GSG98] L. GARRIDO, P, SALEMBIER, and D. GARCIA, Extensive operators in partition latties or image sequence analysis. Signal Processing, 6(2):157-180, 1998.

[KLM94] G. KOEPFLER, C. LOPEZ, and J-M. MOREL, A multiscale algorithm for image segmentation by variational method. SIAM Journal on Numerical Analysis, 31(1):282-299, 1994.

[KSH01] F. KURUGOLLU, B. SANKUR, and A. HARMANCI, Color image segmentation using histogram multithresholding and fusion. Image and Vison Computing, 19(13):915-928, 2001.

[LC02] O. LEZORAY and H. CARDOT, Histogram and watershed based segmentation of color images. In Proceedings of CGIV'2002, pages 358-362, 2002.

[LC03] O. LEZORAY and H. CARDOT, Hybrid color image segmentation using 2d histogram clustering and region merging. In Proceedings of ICISP'2003, volume 1, pages 22-29, 2003.

[LCV02] E. LEFEVRE, O. COLOT, and P. VANNOORENBERGHE, Belief function combination and conflict management, Information Fusion, 3(2) :149-162, 2002.

[Lez04] O. LEZORAY, An unsupervised color image segmentation based on morphological 2d clustering and fusion. In Proceedings of CGIV'2004, 2004.

[LVC00] E. LEFEVRE, P. VANNOORENBERGHE, and O. COLOT, About the use of dempster-shafer theory fot color image segmentation. In Proceedings of CGIP'2000, pages 164-169, 2000.

[LY94] J. LIU and Y-H. YANG, Multiresolution color image segmentation. IEEE transactions on Pattern Analysis and Machine Intelligence, 16(7):689-700, 1994.

[Mac04] L. MACAIRE, Exploitation de la couleur pour la segmentation et l’analyse d’images, Habilitation à diriger des recherches, Université des Sciences et Techniques de Lille, 2004.

[MBVM97] S. MASCLE, I. BLOCH, and D. VIDAL-MADJAR, Application of dempster-shafer evidence theory to unsupervised classification in multisource remote sensing. IEEE Trans. on Geoscience and Remote Sensing, 34(5):1018-1031, 1997.

[MS89] D. MUMFORD and J. SHAH, Optimal approximations by piecewise smooth functions and associated variational problems. Comm. On Pure and Applied Math., 17(4):577-685, 1989.

[PYL98] S.H. PARK, I.D. YUN, and S.U. LEE, Color image segmentation based on 3-d clustering: morphological approach. Pattern Recognition, 31(8):1061-1076, 1998.

[PZLB93] J.G. POSTAIRE, R.D. ZHANG, and C. LECOCQ-BOTTE, Cluster analysis by binary morphology. IEEE Transactions on Pattern Analysis and Machine Intelligence, 15(2):170-180, 1993.

[Ser82] J. SERRA, Image Analysis and mathematical morphology. Academic Press, London, 1982.

[SG00] P. SALEMBIER and L. GARRIDO, Binary partition tree as an efficient representation for image processing, segmentation and information retrieval. IEEE transactions on Image Processing, 9(4):561-576, 2000.

[SGG97] P. SALEMBIER, L. GARRIDO, and D. GARCIA, Image sequence analysis and merging algorithm. In International Workshop on Very Low Bit-rate Video, pages 1-8, 1997.

[Sha76] G. SHAFER, A mathematical theory of evidence. Princeton University Press, 1976.

[Sme90] P. SMETS, Constructing the pignistic probability function in a context of uncertainty. Uncertainty in Artificial Intelligence, 5:29-39, 1990. Elsevier Science Publishers.

[Soi96] P. SOILLE, Morphological partitioning of multispectral images. Journal of Electronic Imaging, 18(4):252-265, 1996.

[Soi04] P. SOILLE, Morphological Image Analysis: Principles and Applications. Springer-Verlag, 2004.

[TFMB04] A. TRÉMEAU, C. FERNANDEZ-MALOIGNE, and P. BONTON, Image numérique couleur. Dunod, 2004.

[XGDL03] H. XUE, T. GÉRAUD, and A. DURET-LUTZ, Multi-band segmentation using morphological clustering and fusion application to color image segmentation. In Proceedings of the IEEE International Conference on Image Processing (ICIP'03), volume 1, pages 353-356, 2003.