Abstract
An approach to handprinted word recognition is described. The approach is based on the use of generating multiple possible segmentations of a word image into characters and matching these segmentations to a lexicon of candidate strings. The segmentation process uses a combination of connected component analysis and distance transform-based, connected character splitting. Neural networks are used to assign character confidence values to potential character within word images. Experimental results are provided for both character and word recognition modules on data extracted from the NIST handprinted character database.
Similar content being viewed by others
References
ATC (1992) Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC
Boccignone G, Chianese A, Cordelia L, Marcelli A (1993) Recovering dynamic information from static handwriting. J Patt Recogn (Special issue on handwriting processing and recognition) 26:409–418
Casey R, Takahashi H (1992) Experience in segmenting and classifying the NIST data base. Proceedings of the 2nd International Workshop on Frontiers in Handwriting Recognition. Chateau de Bonas, France, pp 149–159
Chen M (1992) Off-line handwritten word recognition using hidden Markov models. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 563–579
Chen M, Kundu A (1993) An alternative to variable duration HMM in handwritten word recognition. Proceedings of the 3rd International Workshop of Frontiers in Handwriting Recognition, Buffalo NY, pp 82–92
Favata J, Srihari S (1992) Recognition of general handwritten words using a hypothesis generation and reduction methodology. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 237–253
Gader PD, Whalen MP (1990) Advanced research in handwritten ZIP-code recognition. Technical Report to the United States Postal Service Office of Advanced Technology
Gader PD, HeppD, Forester B, Peurach T, Mitchell B (1990) Pipelined systems for recognition of handwritten digits in USPS ZIP codes. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 539–559
Gader PD, Forester B, Ganzberger M, Gillies A, Mitchell B, Whalen M, Yocum T (1991) Recognition of handwritten digits using template and model matching. J Patt Recogn 24:421–433
Gader P, Mohamed M, Chiang J-H (1992) Segmentation-based handwritten word recognition. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 215–226
Gillies A (1992) Cursive word recognition using hidden Markov models. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 557–563
Gillies AG, Michell BT (1989) A model-based approach to handwritten digit recognition. Machine Vision Appl 2: 231–343
Govindaraju V (1992) Using temporal information in off-line word recognition. Proceedings of the United Stated Postal Service Advanced Technology Conference, Washington DC, pp 529–545
Govindaraju V (1991) Separating handwritten text from overlapping non-textual contours. Proceedings of the 2nd International Workshop on Frontiers in Handwriting Recognition, Chateau de Bonas, France, pp 111–120
Hepp D (1991) An application of backpropagation to the recognition of handwritten digits using morphologically derived features. Proceedings of the 1991 SPIE/SPSE Symposium on Electronic Imaging Science and Technology, Nonlinear Image Processing II Conference, San Jose, Calif.
Hull JJ, Commike A, HO T-K (1990) Multiple algorithms for handwritten character recognition. In: Suen CY (ed) Proceedings of International Workshop on Frontiers in Handwriting Recognition, pp 117–131, Montreal
Hull J, Ho T, Favata J, Govindaraju V, Srihari S (1992) Combination of segmentation-based and holistic handwritten word recogniton algorithms. Proceedings of the 2nd International Workshop on Frontiers in Handwriting Recognition, Chateau de Bonas, France, pp 229–240
IWFHR II (1991) Proceedings of the 2nd International Workshop on Frontiers in Handwriting Recognition, Chateau de Bonas, France
IWFHR III (1993) Proceedings of the 3rd International Workshop on Frontiers in Handwriting Recognition, Buffalo NY
Kumura F, Shridhar M, Tsuruoka S, Chen Z (1992) context-directed handwritten word recognition for postal service applications. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 199–214
Kumura F, Shridhar M, Narasimhamurthi N (1993) Lexicon-directed segmentation — recogniton procedure for unconstrained handwritten words. Prodeedings of the 3rd International Workshop on Frontiers in Handwriting Recognition, Buffalo NY, pp 122–132
Lecolinet E, Crettez J (1991) A grapheme-based segmentation technique for cursive script recognition. Proceedings of the 1st International Conference on Document Analysis and Recognition, Saint Malo, France, pp 740–748
Madhvanath S, Govindaraju V (1992) Using holistic features in handwritten word recognition. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 183–199
Mitchell BT, Gillies AG, Whalen MP, Gader PD (1990) A modelbased computer vision system for the recognition of handwritten address ZIP codes. Technical Report to United States Postal Service Office of Advanced Technology
Nohl C, Burges C, Ben J (1992) Character-based handwritten address word recognition with lexicon. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 167–180
Park H, Lee S (1993) Off-line recognition of large-set handwritten Hangul with hidden Markov models. Proceedings of the 3rd International Workshop on Frontiers in Handwriting Recognition, Buffalo NY, pp 51–62
Plamondon R (1993) Handwriting Processing and recognition. J Patt Recogn (Special Issue on Handwriting Processing and Recognition) 26:379–381
Plessis B (1992) Isolated handwritten word recognition for contextual analysis reading. Proceedings of the United States Postal Service Advanced Technology Conference, Washington DC, pp 579–593
Rabiner L (1989) A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings IEEE 77:257–286
Senior AW, Fallside F (1993) An off-line cursive script recognition system using recurrent error propagation networks. Proceedings of the 3rd Workshop on Frontiers in Handwriting Recognition, Buffalo NY, pp 132–142
Suen C, Nadal C, Mai T, Legault R, Lam L (1990) Recognition of totally unconstrained handwritten numerals based on the concept of multiple experts. In: Suen CY (ed) Proceedings of International Workshop on Frontiers in Handwriting Recognition, pp 131–145
Tappert C, Suen C, Wakahara T (1990) The state-of-the art in on-line handwriting recognition. IEEE Trans Patt Anal Machine Intell 12:787–808
Wilson CL, Garris MD (1990) Handprinted character database. National Institute of Standards Advanced Systems Division Report
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Gader, P., Whalen, M., Ganzberger, M. et al. Handprinted word recognition on a NIST data set. Machine Vis. Apps. 8, 31–40 (1995). https://doi.org/10.1007/BF01213636
Issue Date:
DOI: https://doi.org/10.1007/BF01213636