论文信息 - Ordering of the RGB Space with a Growing Self-organizing Network. Application to Color Mathematical Morphology

Ordering of the RGB Space with a Growing Self-organizing Network. Application to Color Mathematical Morphology

Mathematical morphology is broadly used in image processing, but it is mainly restricted to binary or greyscale images. Extension to color images is not straightforward due to the need of application to an ordered space with an infimum and a supremum. In this paper a new approach for the ordering of the RGB space is presented. The adaptation of a linear growing self-organizing network to the three-dimensional color space allows the definition of an order relationship among colors. This adaptation is measured with the topographic product to guarantee a good topology-preservation of the RGB space. Once an order has been established, several examples of application of mathematical morphology operations to color images are presented.

Francisco Flórez-Revuelta | Francisco Flórez-Revuelta

[1] Allan Hanbury,et al. Mathematical Morphology in the HLS Colour Space , 2001, BMVC.

[2] Yeuvo Jphonen,et al. Self-Organizing Maps , 1995 .

[3] J. Chanussot,et al. Total ordering based on space filling curves for multivalued morphology , 1998 .

[4] V. Barnett. The Ordering of Multivariate Data , 1976 .

[5] Bernd Fritzke,et al. Growing cell structures--A self-organizing network for unsupervised and supervised learning , 1994, Neural Networks.

[6] Richard Alan Peters,et al. Mathematical morphology for angle-valued images , 1997, Electronic Imaging.

[7] M. V. Velzen,et al. Self-organizing maps , 2007 .

[8] Edward J. Delp,et al. Morphological operations for color image processing , 1999, J. Electronic Imaging.

[9] Klaus Pawelzik,et al. Quantifying the neighborhood preservation of self-organizing feature maps , 1992, IEEE Trans. Neural Networks.