A computer experiment in pattern theory
- 1 January 1989
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics. Stochastic Models
- Vol. 5 (4) , 531-553
- https://doi.org/10.1080/15326348908807123
Abstract
A pattern theoretic model of random shape is described and applied to noisy digital images of human hand. Algorithms are derived from the model for processing the pictures, in particular for image restoration. To learn about the power and limitations of the techniques we introduce more and more complex visual noise until the algorithms fail. The noise model is then modified in order to compensate for the peculiarities of the noise. In order to make the techniques computationally feasible analytical results have been exploited to speed up stochastic relaxation. For massive mathematical experiments on the computer, such as the present one, the mathematician needs more advanced computer resources than are usually available. The need for this is emphasized, in particular for powerful single user machines, and appropriate support personnelKeywords
This publication has 6 references indexed in Scilit:
- A limit theorem for pattern synthesis in image processingJournal of Multivariate Analysis, 1990
- Statistics, images, and pattern recognitionThe Canadian Journal of Statistics / La Revue Canadienne de Statistique, 1986
- Nonstationary Markov chains and convergence of the annealing algorithmJournal of Statistical Physics, 1985
- Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of ImagesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1984
- Regular StructuresPublished by Springer Nature ,1981
- Structural Pattern RecognitionPublished by Springer Nature ,1977