We present a novel statistical and variational approach to image segmentation based on a new algorithm named region competition. This algorithm is derived by minimizing a generalized Bayes/MDL(Minimum Description Length) criterion using the variational principle, and it combines the best features of snake/balloon model and region growing. Thus we show that existing techniques in early vision such as, Snake/balloon models, region growing, and Bayes/MDL are addressing different views of the same problem and they can be unified within a common statistical framework to gain the advantage of all of them. We further analyze how to optimize the precision of the resulting boundary location by studying the statistical properties of the region competition algorithm, and discuss what are the "good " initial conditions for the algorithm. We will briefly discuss the relations between region competition and local edge filters. We also present a novel color model, and apply our method to color and texture segmentation. Our algorithm is demonstrated
|
2322
|
Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images
– Geman, Geman
- 1984
|
|
1943
|
Snakes: Active contour models
– Kass, Witkin, et al.
- 1987
|
|
1663
|
Introduction to Statistical Pattern Recognition
– Fukunaga
- 1990
|
|
1662
|
A computational approach to edge detection
– Canny
- 1986
|
|
536
|
Visual Reconstruction
– Blake, Zisserman
- 1987
|
|
433
|
Optimal approximations by piecewise smooth functions and associated variational problems
– Mumford, Shah
- 1989
|
|
315
|
On active contour models and balloons
– Cohen
- 1991
|
|
209
|
Seeded region growing
– Adams, Bischof
- 1994
|
|
188
|
Constructing simple stable description for image partitioning
– Leclerc
- 1989
|
|
133
|
Filters, random fields and maximum entropy (frame
– Zhu, Wu, et al.
- 1998
|
|
130
|
A physical approach to color image understanding,” in
– Klinker, Shafer, et al.
- 1993
|
|
90
|
Forms: A flexible object recognition and modeling system
– Zhu, Yuille
- 1996
|
|
87
|
Region-Based Strategies for Active Contour Models,” Int
– Ronfard
- 1994
|
|
81
|
Automatic extraction of deformable part models
– Pentland
- 1990
|
|
52
|
Segmenting Images Using Normalized Color
– Healey
- 1992
|
|
52
|
Segmentation of range images as the search for geometric parametric models
– Leonardis, Gupta, et al.
- 1995
|
|
50
|
A multiscale algorithm for image segmentation by variational method
– Koepfler, Lopez, et al.
- 1994
|
|
48
|
Another interpretation of the EM algorithm for mixture distribution
– Hathaway
- 1986
|
|
43
|
A finite element method applied to new active contour models and 3D reconstruction from cross sections
– Cohen, Cohen
- 1990
|
|
27
|
The uncertainty principle in image processing
– Wilson, Granlund
- 1984
|
|
26
|
Deformable boundary finding influenced by region homogeneity
– Chakraborty, Staib, et al.
- 1994
|
|
26
|
Compact region extraction using weighted pixel linking in a pyramid
– Hong, Rosenfeld
|
|
16
|
Robust Active Contours with Insensitive Parameters
– Xu, Segawa, et al.
- 1994
|
|
13
|
Modeling by Shortest Data Description," Automatica 14
– Rissanen
- 1978
|
|
10
|
Foundations of Applied Mathematics
– Greenberg
- 1978
|
|
7
|
A Vector Signal Processing Approach to Color
– Sung
- 1992
|
|
6
|
Modeling light reflection for color computer vision
– Lee, Breneman, et al.
- 1990
|
|
5
|
Map representations and coding-based priors for segmentation
– Keeler
- 1991
|
|
5
|
O(log n) Bimodality Analysis
– Phillips, Rosenfeld, et al.
- 1989
|
|
2
|
A.L.Yuille. "A common framework for image segmentation
– Geiger
- 1991
|
|
2
|
P.Dong. "Boundary Detection by Constrained Optimization
– Geman
- 1990
|
|
2
|
The Heat Equation Shrinking Convex Plane Curves
– Grayson
- 1987
|
|
2
|
Applied Multivariate Statsitical Analysis
– Johnson, Wichem
- 1982
|
|
2
|
B.Dom, W.Niblack and D.Steele. "A fast algorithm for MDL-based multi-band image segmentation
– Kanungo
- 1994
|
|
2
|
A.P.Pentland, "Cooperative Robust Estimation Using Layers of Support
– Darrell
- 1994
|
|
1
|
Existence and Regularity of Solutions to a Variational Problem of Mumford and Shah
– Wang
- 1991
|
|
1
|
Learning and sampling the prior distribution for visual computation". Harvard Robotics Laboratory
– Zhu, Mumford
- 1995
|
|
1
|
A Unified Theory for Image Segmenation: Region Competition and its Analysis". Harvard Robotics Laboratory
– Zhu, Yuille
- 1995
|
|
