Bidirectional Long-Short Term Memory Based Recurrent Neural Network for Handwriting Recognition

International Journal of Computer Science (IJCS) Published by SK Research Group of Companies (SKRGC)

Abstract

Today, BLSTM is widely used in recurrent neural network for speech and handwriting recognition. In this present paper a novel type of recurrent neural network is specifically designed for sequence labeling tasks where the data is hard to segment and contains long-range bidirectional interdependencies. For many tasks, BLSTM is useful to have access to future, as well as past, context. We refer keyword spotting method for handwritten documents. It is derived from a neural network-based system for unconstrained handwriting recognition. As such it performs template-free spotting, i.e., it is not necessary for a keyword to appear in the training set. The keyword spotting is done using a modification of the CTC Token Passing algorithm in conjunction with a recurrent neural network and a statistical n-gram language model instead of a bigram language model.

References

[1] Volkmar Frinken, Andreas Fischer, R. Manmatha and Horst Bunke, “A Novel Word Spotting Method Based on Recurrent Neural Networks”, IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 33, no. 4 feb, 2012.

[2] Alex Graves, Marcus Liwicki, Santiago Fernandez, Roman Bertolami, Horst Bunke, and Jurgen Schmidhuber, “A Novel Connectionist System for Unconstrained Handwriting Recognition” , IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 31,no. 5,feb 2009

[3] A. Vinciarelli, “A Survey on Off-Line Cursive Word Recognition”, Pattern Recognition, vol. 35, no. 7, pp. 1433-1446, 2002.

[4] R. Plamondon and S.N. Srihari, “On-Line and O-Line Hand-writing Recognition: A Comprehensive Survey”, IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 22, no. 1, pp. 63-84,Jan. 2000.

[5] S.-S. Kuo and O.E. Agazzi, “Keyword Spotting in Poorly Printed Documents Using Pseudo 2-D Hidden Markov Models”, IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 16, no. 8, pp. 842-848, Aug. 1994.

[6] R. Manmatha, C. Han, and E. Riseman, “Word Spotting : A New Approach to Indexing Handwriting” , Proc. IEEE Int'l Conf. Computer Vision and Pattern Recognition, pp.631-637, 1996.

[7] U.-V. Marti and H. Bunke, “Using a Statistical Language Model to Improve the Performance of an HMM-Based Cursive Hand-writing Recognition System,” Int’l J. Pattern Recognition and Artificial Intelligence, vol. 15, pp. 65-90, 2001.

[8] A. Bhardwaj, D. Jose, and V. Govindaraju, “Script Independent Word Spotting in Multilingual Documents” , Proc. Second Int'l Workshop Cross Lingual Information Access,pp. 48-54, 2008.

[9] Y. Leydier, F. Lebourgeois, and H. Emptoz , “Text Search for Medieval Manuscript Images” , Pattern Recognition, vol. 40, pp. 3552-3567, 2007.

[10] A. Kotcz, J. Alspector, M.F. Augusteijn, R. Carlson, and G.V. Popescu, “A Line-Oriented Approach to Word Spotting in Handwritten Documents” , Pattern Analysis and Applications, vol. 3, pp. 153-168, 2000.

[11] H. Cao, A. Bhardwaj, and V. Govindaraju, “A Probabilistic Method for Keyword Retrieval in Handwritten Document Images” , Pattern Recognition, vol. 42, no. 12, pp 3374-3382, https://dx.doi.org/10.1016/j.patcog.2009.02.003, Dec. 2009.

[12] J. Chan, C. Ziftci, and D. Forsyth, “Searching Off-Line Arabic Documents” , Proc. IEEE Conf. Computer Vision and Pattern Recognition, pp. 1455-1462, 2006.

[13] E. Saykol, A.K. Sinop, U. Güdükbay, O. Ulusoy, and A.E. Cetin, “Content-Based Retrieval of Historical Ottoman Documents Stored as Textual Images” , IEEE Trans. Image Processing, vol. 13, no. 3, pp. 314-325, Mar. 2004.

[14] R.F. Moghaddam and M. Cheriet, “Application on Multi-Level Classifier and Clustering for Automatic Word Spotting in Historical Document Images” , Proc. 10th Int'l Conf. Document Analysis and Recognition, vol. 2, pp. 511-515, 2009.

Keywords

Keyword spotting, offline handwriting, document analysis, historical documents, neural network, BLSTM.