TY - GEN
T1 - Improved BLSTM neural networks for recognition of on-line bangla complex words
AU - Frinken, Volkmar
AU - Bhattacharya, Nilanjana
AU - Uchida, Seiichi
AU - Pal, Umapada
PY - 2014
Y1 - 2014
N2 - While bi-directional long short-term (BLSTM) neural network have been demonstrated to perform very well for English or Arabic, the huge number of different output classes (characters) encountered in many Asian fonts, poses a severe challenge. In this work we investigate different encoding schemes of Bangla compound characters and compare the recognition accuracies. We propose to model complex characters not as unique symbols, which are represented by individual nodes in the output layer. Instead, we exploit the property of long-distance-dependent classification in BLSTM neural networks. We classify only basic strokes and use special nodes which react to semantic changes in the writing, i.e., distinguishing inter-character spaces from intra-character spaces. We show that our approach outperforms the common approaches to BLSTM neural network-based handwriting recognition considerably.
AB - While bi-directional long short-term (BLSTM) neural network have been demonstrated to perform very well for English or Arabic, the huge number of different output classes (characters) encountered in many Asian fonts, poses a severe challenge. In this work we investigate different encoding schemes of Bangla compound characters and compare the recognition accuracies. We propose to model complex characters not as unique symbols, which are represented by individual nodes in the output layer. Instead, we exploit the property of long-distance-dependent classification in BLSTM neural networks. We classify only basic strokes and use special nodes which react to semantic changes in the writing, i.e., distinguishing inter-character spaces from intra-character spaces. We show that our approach outperforms the common approaches to BLSTM neural network-based handwriting recognition considerably.
UR - http://www.scopus.com/inward/record.url?scp=84906309034&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84906309034&partnerID=8YFLogxK
U2 - 10.1007/978-3-662-44415-3_41
DO - 10.1007/978-3-662-44415-3_41
M3 - Conference contribution
AN - SCOPUS:84906309034
SN - 9783662444146
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 404
EP - 413
BT - Structural, Syntactic, and Statistical Pattern Recognition - Joint IAPR International Workshop, S+SSPR 2014, Proceedings
PB - Springer Verlag
T2 - Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition, S+SSPR 2014
Y2 - 20 August 2014 through 22 August 2014
ER -