Abstract
Relaxation labeling processes are a class of parallel iterative procedures widely used in artificial intelligence and computer vision. Recently, a learning algorithm for relaxation labeling has been developed which involves minimizing a certain cost function with a gradient method. Despite the encouraging results obtained so far, the gradient algorithm suffers from some drawbacks that could prevent its application to practical problems. Essentially, these include the inability to escape from local minima and its computational complexity. In this paper we attempt to overcome the difficulties with the gradient procedure and propose the use of genetic algorithms for solving the learning problem of relaxation. Some results are reported which prove the effectiveness of the proposed approach.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
R. M. Haralick and L. G. Shapiro (1979). The consistent labeling problem: Part I. IEEE Trans. Pattern Anal. Machine Intell. 1(2), 173–184.
E. Charniak and D. McDermott (1985). Introduction to Artificial Intelligence. Reading, MA: Addison-Wesley.
A. Rosenfeld, R. A. Hummel and S. W. Zucker (1976). Scene labeling by relaxation operations. IEEE Trans. Syst. Man Cybern. 6(6), 420–433.
L. S. Davis and A. Rosenfeld (1981). Cooperating processes for low-level vision: A survey. Artificial Intell. 17, 245–263.
R. A. Hummel and S. W. Zucker (1983). On the foundations of relaxation labeling processes. IEEE Trans. Pattern Anal. Machine Intell. 5(3), 267–287.
J. Kittler and J. Illingworth (1985). Relaxation labeling algorithms — A review. Image Vision Comput. 3(4), 206–216.
R. M. Haralick, J. L. Mohammed and S. W. Zucker (1980). Compatibilities and the fixed points of arithmetic relaxation processes. Comput. Graph. Image Processing 13, 242–256.
D. P. O'Leary and S. Peleg (1983). Analysis of relaxation processes: The two-node two-label case. IEEE Trans. Syst. Man Cybern. 13(4), 618–623.
S. Peleg and A. Rosenfeld (1978). Determining compatibility coefficients for curve enhancement relaxation processes. IEEE Trans. Syst. Man Cybern. 8(7), 548–555.
M. Pelillo and M. Refice (1992). An optimization algorithm for determining the compatibility coefficients of relaxation labeling processes. Proc. 11th Int. Conf. Pattern Recognition, The Hague, The Netherlands, 145–148.
M. Pelillo and M. Refice (submitted). Learning compatibility coefficients for relaxation labeling processes. IEEE Trans. Pattern Anal. Machine Intell.
S. W. Zucker, Y. G. Leclerc and J. L. Mohammed (1981). Continuous relaxation and local maxima selection: Conditions for equivalence. IEEE Trans. Pattern Anal. Machine Intell. 3(3), 117–127.
D. E. Goldberg (1989). Genetic Algorithms in Search, Optimization, and Machine Learning. Reading, MA: Addison-Wesley.
J. H. Holland (1992). Adaptation in Natural and Artificial Systems. 2nd ed., Cambridge, MA: MIT Press.
S. Kullback (1959). Information Theory and Statistics. New York: Wiley.
J. Lin (1991). Divergence measures based on the Shannon entropy. IEEE Trans. Inform. Theory 37(1), 145–151.
T. P. Caudell and C. P. Dolan (1989). Parametric connectivity: Training of constrained networks using genetic algorithms. Proc. 3rd Int. Conf. Genetic Algorithms, George Mason University, 370–374.
K. W. Church (1989). A stochastic parts program and noun phrase parser for unrestricted text. Proc. IEEE Int. Conf. Acoustics Speech Signal Processing, Glasgow, Scotland, 695–698.
M. Pelillo and M. Refice (1991). Syntactic category disambiguation through relaxation processes. Proc. EUROSPEECH 91, Genova, Italy, 757–760.
M. Pelillo, F. Abbattista and N. Abbattista (1993). Globally optimal learning for relaxation labeling by simulated annealing. Proc. 7th Int. Conf. Image Anal. Processing, Bari, Italy.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1993 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Pelillo, M., Abbattista, F., Maffione, A. (1993). Evolutionary learning for relaxation labeling processes. In: Torasso, P. (eds) Advances in Artificial Intelligence. AI*IA 1993. Lecture Notes in Computer Science, vol 728. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-57292-9_61
Download citation
DOI: https://doi.org/10.1007/3-540-57292-9_61
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-57292-3
Online ISBN: 978-3-540-48038-9
eBook Packages: Springer Book Archive