Abstract
Traditional approaches to combining classifiers attempt to improve classification accuracy at the cost of increased processing. They may be viewed as providing an accuracy-speed trade-off: higher accuracy for lower speed. In this paper we present a novel approach to combining multiple classifiers to solve the inverse problem of significantly improving classification speeds at the cost of slightly reduced classification accuracy. We propose a cascade architecture for combining classifiers and cast the process of building such a cascade as a search and optimization problem. We present two algorithms based on steepest-descent and dynamic programming for producing approximate solutions fast. We also present a simulated annealing algorithm and a depth-first-search algorithm for finding optimal solutions. Results on handwritten optical character recognition indicate that a) a speedup of 4-9 times is possible with no increase in error and b) speedups of up to 15 times are possible when twice as many errors can be tolerated.
Chapter PDF
Similar content being viewed by others
Keywords
- Simulated Annealing
- Hide Node
- Simulated Annealing Algorithm
- Multiple Classifier
- Convolutional Neural Network
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
References
Ludmila, I.K.: A Theoretical Study on Six Classifier Fusion Strategies. IEEE Trans. On Pattern Analysis and Machine Intelligence 24(2), 281–286 (2002)
Duin, R.: The Combining Classifier: To Train or Not to Train? ICPR (2), 765–770 (2002)
Kittler, J., Hatef, M., Duin, R.P.W., Matas, J.: On Combining Classifiers. IEEE Trans. On Pat. Analysis and Machine Intel. 20(3) (March 1998)
Prevost, L., Michel-Sendis, C., Moises, A., Oudot, L., Milgram, M.: Combining model-based and discriminative classifiers: application to handwritten character recognition. In: ICDAR 2003 (2003)
Hao, H., Liu, C.L., Sako, H.: Confidence evaluation for combining diverse classifiers. In: ICDAR 2003, August 3-6, pp. 760–765 (2003)
Bhattacharya, U., Chaudhuri, B.B.: A Majority Voting Scheme for Multiresolution Recognition of Handprinted Numerals. In: ICDAR 2003, August 3-6, pp. 16–20 (2003)
Simard, P.Y., Steinkraus, D., Platt, J.: Best Practice for Convolutional Neural Networks Applied to Visual Document Analysis. In: ICDAR 2003, pp. 958–962 (2003)
Marinai, S., Gori, M., Soda, G.: Artificial Neural Networks for Document Analysis and Recognition. IEEE TPAMI 27(1), 23–35
Zhang, T., Yu, B.: Boosting with early stopping: Convergence and consistency. Annals of Statistics 33(4), 1538–1579 (2005)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2006 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Chellapilla, K., Shilman, M., Simard, P. (2006). Combining Multiple Classifiers for Faster Optical Character Recognition. In: Bunke, H., Spitz, A.L. (eds) Document Analysis Systems VII. DAS 2006. Lecture Notes in Computer Science, vol 3872. Springer, Berlin, Heidelberg. https://doi.org/10.1007/11669487_32
Download citation
DOI: https://doi.org/10.1007/11669487_32
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-32140-8
Online ISBN: 978-3-540-32157-6
eBook Packages: Computer ScienceComputer Science (R0)