Three-dimensional morphological pyramid and its application to color image sequence coding

doi:10.1016/0165-1684(95)00022-6

Signal Processing

Volume 44, Issue 2, June 1995, Pages 163-180

https://doi.org/10.1016/0165-1684(95)00022-6 Get rights and content

Abstract

In this paper, we propose an efficient and multiresolutional image sequence representation scheme in which the input image sequence is decomposed and represented as three-dimensional (3-D) morphological pyramids. An input image sequence is first decomposed into a 3-D morphological lowpass pyramid using 3-D morphological filters. A 3-D morphological bandpass pyramid is then built by taking the difference between the parent and the expanded son or daughter. Finally, a 3-D vector quantization is applied to each level of the 3-D morphological bandpass pyramid to reduce the bandwidth required for transmitting the bandpass pyramid. An example of color image sequence is presented to show the effectiveness of this coding system.

Zusammenfassung

Dieser Beitrag stellt ein Verfahren zur effizienten und mehrfachauflösenden Beschreibung von Bildsequenzen vor. Die Bildsequenz wird mit Hilfe dreidimensionaler (3-D) morphologischer Filter in eine morphologische (3-D) Pyramide zerlegt. Einer Zerlegung in Tiefpass-Signale folgt die Zerlegung in eine Bandpasspyramide mittels Komplementbildung zwischen aufeinanderfolgenden Hierarchieebenen. Schlieβlich wird auf jeder Ebene der (3-D) morphologischen Bandpasspyramide eine Vektor-Quantiesierung durchgeführt, um so die zur Übertragung der Bandpasspyramide erforderliche Bandbreite zu reduzieren. Die Wirksamkeit dieses Kodierungsverfahrens wird anhand eine Beispiels mit Farbbildsequenzen verdeutlicht.

Résumé

Nous proposons dans cet article une technique de représentation de séquences d'images multi-résolution efficace dans laquelle la séquence d'images d'entrée est décomposée en et représentée par des pyramides morphologiques tridimensionnelles (3-D). Une séquence d'images est tout d'abord décomposée en une pyramide passe-bas morphologique 3-D à l'aide de filtres morphologiques 3-D. Une pyramide morphologique passe-bande 3-D est ensuite construite en prenant la différence entre la séquence de base et la séquence produite. Enfin, une technique de quantification vectorielle 3-D est appliquée à chaque niveau de la pyramide morphologique passe-band 3-D afin de réduire la largeur de bande requise pour la transmission de celle-ci. Un exemple sur une séquence d'images couleur est présenté pour mettre en évidence l'efficacité de ce système de codage.

References (29)

R.-J. Chen et al.
A 3-dimensional morphological image sequence coding
S.R. Sternberg
Graylevel morphology
CVGIP
(1986)
A. Toet
A morphological pyramidal image decomposition
Pattern Recognition Lett.
(May 1989)
P.J. Burt et al.
The Laplacian pyramid as a compact image code
IEEE Trans. Commun.
(April 1983)
J. Foster et al.
Finite-state vector quantization for waveform coding
IEEE Trans. Inform. Theory
(May 1985)
M. Goldberg et al.
Image sequence coding using vector quantization
IEEE Trans. Commun.
(July 1986)
M. Goldberg et al.
Comparative performance of pyramid data structures for progressive image transmission
IEEE Trans. Commun.
(April 1991)
R.M. Gray
Vector quantization
IEEE ASSP Mag.
(April 1984)
R.M. Haralick et al.
Image analysis using mathematical morphology
IEEE Trans. Pattern Anal. Mach. Intell.
(July 1987)
R.M. Haralick et al.
The digital morphological sampling theorem

N.S. Jayant et al.

DIGITAL CODING OF WAVE-FORMS: Principles and Applications to Speech and Video

P. Maragos

Tutorial on advances in morphological image processing and analysis

Optical Eng.

(July 1987)

P.A. Maragos et al.

Morphological skeleton representation and coding of binary images

IEEE Trans. Acoust. Speech Signal Process.

(October 1986)

P. Maragos et al.

Morphological filters-Part I: Their set-theoretic analysis and relations to linear shift-invariant filters

IEEE Trans. Acoust. Speech Signal Process.

(August 1987)

Cited by (4)

Near real-time Evolution-based Adaptation Strategy for dynamic Color Quantization of image sequences
2000, Information sciences
Citation Excerpt :
Color Quantization of images within a sequence contains the essence of the paradigm of real time Non-stationary Clustering. Although sequences of images (video) lead naturally to the consideration of time varying Clustering/VQ problems, the usual approaches to the computation of codebooks for both Color Quantization and VQ of image sequences consider time invariant distributions of colors [19] or image blocks [17], and apply conventional Clustering methods. Some heuristic efforts [18,27] have been reported that try to cope with the time varying characteristics inherent to image sequences.
Color Quantization of still images can be easily stated as a Clustering problem. Color Quantization of sequences of images becomes a Non-stationary Clustering Problem. In this paper we propose a very simple and effective Evolution-based Adaptive Strategy to perform the adaptive computation of the color representatives for each image in the sequence. Salient features of the algorithm proposed here are: individuals correspond to individual cluster centers, to approach real-time response we impose one-generation adaptation for each image, mutation operators are guided by the actual covariance matrices of the clusters. Experimental results on a sequence of indoor images are presented.
Colour morphological scale-spaces from the positional colour sieve
2005, Proceedings of the Digital Imaging Computing: Techniques and Applications, DICTA 2005
Basic competitive neural networks as adaptive mechanisms for non-stationary colour quantisation
1999, Neural Computing and Applications
Competitive neural networks as adaptive algorithms for nonstationary clustering: Experimental results on the color quantization of image sequences
1997, IEEE International Conference on Neural Networks - Conference Proceedings

View full text

PaperThree-dimensional morphological pyramid and its application to color image sequence coding

Abstract

Zusammenfassung

Résumé

CVGIP

Pattern Recognition Lett.

The Laplacian pyramid as a compact image code

IEEE Trans. Commun.

Finite-state vector quantization for waveform coding

IEEE Trans. Inform. Theory

Image sequence coding using vector quantization

IEEE Trans. Commun.

Comparative performance of pyramid data structures for progressive image transmission

IEEE Trans. Commun.

Vector quantization

IEEE ASSP Mag.

Image analysis using mathematical morphology

IEEE Trans. Pattern Anal. Mach. Intell.

The digital morphological sampling theorem

DIGITAL CODING OF WAVE-FORMS: Principles and Applications to Speech and Video

Tutorial on advances in morphological image processing and analysis

Optical Eng.

Morphological skeleton representation and coding of binary images

IEEE Trans. Acoust. Speech Signal Process.

Morphological filters-Part I: Their set-theoretic analysis and relations to linear shift-invariant filters

IEEE Trans. Acoust. Speech Signal Process.

Paper
Three-dimensional morphological pyramid and its application to color image sequence coding