Abstract
A recognition procedure is modelled as a queuing problem. We obtain first the steady-state distribution of probabilities of the basic process describing its evolution. Then, we derive some performance measures of the recognition procedure such as the waste time, the queue length of patterns, the classification error probabilities and so on. Stochastic comparisons are also provided. Our study extend the Viscolani's results to the non Markovian case.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Alzaid A., J.S. Kim, F. Proshan, (1991). Laplace ordering and its applications,J. Appl. Prob. 28, 116–130.
Bendrissou B., E. Goutsouliak, (1987). Modèle stochastique de classification en reconnaissance des formes, ler colloque Maghrébin sur lesModèles Numériques de l'ingénieur, Alger, vol.1, 1–9.
Fu, K.S. (1968).Sequential Methods in Pattern Recognition and Machine Learning, Academic Press, New York.
Gnedenko B.V., Kovalenko I.N., (1965). Introduction to Queuing Theory,Nauka, Moscow (in russian), Englis, translat:Weiner Bindery, (1968).
Stoyan D. (1983).Comparison Methods for Queues and other Stochastic Models, Wiley, New York.
Viscolani, B., (1984). A Queuing Problem with Pattern Recognition,RAIRO Recherche Opérationnelle,18–1, 71–88.
Author information
Authors and Affiliations
Additional information
An erratum to this article is available at http://dx.doi.org/10.1007/BF02564833.
Rights and permissions
About this article
Cite this article
Aissani, A., de Beauville, J.P.A. Queuing analysis of a sequential recognizer of patterns. Top 6, 45–66 (1998). https://doi.org/10.1007/BF02564798
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/BF02564798
Key Words
- Pattern recognition
- Performance evaluation
- Queuing system
- Stationary distribution
- Classification error probabilities
- Stochastic comparison