Handwritten isolated word recognition: an approach based on Mutual Information for feature set validation

The paper presents the application of the Mutual Information criterion (T.M. Cover and J.A. Thomas, 1991) to validate feature sets extracted from handwritten words in Brazilian legal amounts. The lexicon includes a subset of short words without ascenders/descenders and subsets of words with the same prefix or suffix. These particularities of the Brazilian lexicon show that it is necessary to improve the perpetual feature set with complementary geometric features, and also modeling the prefix and suffix of the words. Finally, the experiments show the viability of our approach.

[1]  Daniel Ollivier Une approche economisant les traitements pour reconnaitre l'ecriture manuscrite : application a la reconnaissance des montants litteraux de cheques bancaires , 1999 .

[2]  Manuel Avila Optimisation de modèles markoviens pour la reconnaissance de l'écrit , 1996 .

[3]  Nicholas W Strathy A method for segmentation of touching handwritten numerals , 1993 .

[4]  李幼升,et al.  Ph , 1989 .

[5]  Robert Sabourin,et al.  Isolated word recognition in brazilian bank check legal amounts , 2000 .

[6]  Jim R. Parker,et al.  Algorithms for image processing and computer vision , 1996 .

[7]  Frédéric Grandidier,et al.  Une nouvelle stratégie pour l'amélioration des jeux de primitives d'un système de reconnaissance de l'écriture , 2000 .

[8]  Venu Govindaraju,et al.  Perceptual Features for Off-line Handwritten Word Recognition: A Framework for Heuristic Prediction, Representation and Matching , 1998, SSPR/SPR.

[9]  Lambertus Schomaker A method for the determination of features used in human reading of cursive handwriting. , 1998 .

[10]  Robert Sabourin,et al.  An HMM-Based Approach for Off-Line Unconstrained Handwritten Word Modeling and Recognition , 1999, IEEE Trans. Pattern Anal. Mach. Intell..

[11]  Roberto Battiti,et al.  Using mutual information for selecting features in supervised neural net learning , 1994, IEEE Trans. Neural Networks.

[12]  Eric Lecolinet Segmentation d'images de mots manuscrits : application a la lecture de chaines de caracteres majuscules alphanumeriques et a la lecture de l'ecriture cursive , 1990 .

[13]  Thomas M. Cover,et al.  Elements of Information Theory , 2005 .

[14]  Biing-Hwang Juang,et al.  Fundamentals of speech recognition , 1993, Prentice Hall signal processing series.

[15]  Natanael Rodrigues Gomes,et al.  Reconhecimemto de palavras manuscritas baseado em HMM e no emprego de caracteristicas topologicas e geometricas , 2000 .

[16]  Gyeonghwan Kim,et al.  Recognition of offline handwritten words and its extension to phrase recognition , 1996 .

[17]  Didier Guillevic Unconstrained handwriting recognition applied to the processing of bank cheques , 1996 .

[18]  Cinthia Obladen de Almendra Freitas,et al.  Brazilian bank check handwritten legal amount recognition , 2000, Proceedings 13th Brazilian Symposium on Computer Graphics and Image Processing (Cat. No.PR00878).