by Tamas Sziranyi, Josiane Zerubia, Laszlo Czuni, David Geldreich, Zoltan Kato
Abstract:
Markovian approaches to early vision processes need a huge amount of computing power. These algorithms can usually be implemented on parallel computing structures. Herein, we show that the Markovian labeling approach can be implemented in fully parallel cellular network architectures, using simple functions and data representations. This makes possible to implement our model in parallel imaging VLSI chips. As an example, we have developed a simplified statistical image segmentation algorithm for the Cellular Neural/Nonlinear Networks Universal Machine (CNN-UM), which is a new image processing tool, containing thousands of cells with analog dynamics, local memories and processing units. The Modified Metropolis Dynamics (MMD) optimization method can be implemented into the raw analog architecture of the CNN-UM. We can introduce the whole pseudo-stochastic segmentation process in the CNN architecture using 8 memories/cell. We use simple arithmetic functions (addition, multiplication), equality-test between neighboring pixels and very simple nonlinear output functions (step, jigsaw). With this architecture, the proposed VLSI CNN chip can execute a pseudo-stochastic relaxation algorithm of about 100 iterations in about 100 s. In the suggested solution the segmentation is unsupervised, where a pixel-level statistical estimation model is used. We have tested different monogrid and multigrid architectures. In our CNN-UM model several complex preprocessing steps can be involved, such as texture-classification or anisotropic diffusion. With these preprocessing steps, our fully parallel cellular system may work as a high-level image segmentation machine, using only simple functions based on the close-neighborhood of a pixel.
Reference:
Tamas Sziranyi, Josiane Zerubia, Laszlo Czuni, David Geldreich, Zoltan Kato, Image Segmentation Using Markov Random Field Model in Fully Parallel Cellular Network Architectures, In Real Time Imaging, volume 6, no. 3, pp. 195-211, 2000.
Bibtex Entry:
@string{rti="Real Time Imaging"}
@Article{Sziranyi-etal2000,
author = {Sziranyi, Tamas and Zerubia, Josiane and Czuni,
Laszlo and Geldreich, David and Kato, Zoltan},
title = {Image Segmentation Using {M}arkov Random Field Model
in Fully Parallel Cellular Network Architectures},
journal = rti,
year = 2000,
volume = 6,
number = 3,
pages = {195--211},
month = jun,
pdf = {papers/rti2000.pdf},
abstract = {Markovian approaches to early vision processes need
a huge amount of computing power. These algorithms
can usually be implemented on parallel computing
structures. Herein, we show that the Markovian
labeling approach can be implemented in fully
parallel cellular network architectures, using
simple functions and data representations. This
makes possible to implement our model in parallel
imaging VLSI chips. As an example, we have developed
a simplified statistical image segmentation
algorithm for the Cellular Neural/Nonlinear Networks
Universal Machine (CNN-UM), which is a new image
processing tool, containing thousands of cells with
analog dynamics, local memories and processing
units. The Modified Metropolis Dynamics (MMD)
optimization method can be implemented into the raw
analog architecture of the CNN-UM. We can introduce
the whole pseudo-stochastic segmentation process in
the CNN architecture using 8 memories/cell. We use
simple arithmetic functions (addition,
multiplication), equality-test between neighboring
pixels and very simple nonlinear output functions
(step, jigsaw). With this architecture, the proposed
VLSI CNN chip can execute a pseudo-stochastic
relaxation algorithm of about 100 iterations in
about 100 s. In the suggested solution the
segmentation is unsupervised, where a pixel-level
statistical estimation model is used. We have tested
different monogrid and multigrid architectures. In
our CNN-UM model several complex preprocessing steps
can be involved, such as texture-classification or
anisotropic diffusion. With these preprocessing
steps, our fully parallel cellular system may work
as a high-level image segmentation machine, using
only simple functions based on the
close-neighborhood of a pixel.}
}