论文信息 - Estimation of prior probabilities for numeral recognition

Estimation of prior probabilities for numeral recognition

Abstract An approach for estimating a priori probabilities of numerals is proposed. It is based on a special approximation of a training set of histograms. An application of the approach to recognition of check amounts is considered. Experimental results with real bank check databases are presented.

N. D. Gorsky | Nickolas D. Gorsky | L. M. Gorskaya

[1] Michel Loève,et al. Probability Theory I , 1977 .

[2] Françoise Fogelman-Soulié,et al. Neurocomputing : algorithms, architectures and applications , 1990 .

[3] Jack Perkins,et al. Pattern recognition in practice , 1980 .

[4] Nikolai Gorski,et al. Bank Check Reading: Recognizing the Courtesy Amount , 1995, ICSC.

[5] Frederick Jelinek,et al. Interpolated estimation of Markov source parameters from sparse data , 1980 .

[6] Michel Gilloux,et al. Strategies for handwritten words recognition using hidden Markov models , 1993, Proceedings of 2nd International Conference on Document Analysis and Recognition (ICDAR '93).

[7] Richard O. Duda,et al. Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[8] John Scott Bridle,et al. Probabilistic Interpretation of Feedforward Classification Network Outputs, with Relationships to Statistical Pattern Recognition , 1989, NATO Neurocomputing.

[9] Arthur Nádas,et al. On Turing's formula for word probabilities , 1985, IEEE Trans. Acoust. Speech Signal Process..