TY - GEN
T1 - Improving handwritten Chinese text recognition by confidence transformation
AU - Wang, Qiu Feng
AU - Yin, Fei
AU - Liu, Cheng Lin
PY - 2011
Y1 - 2011
N2 - This paper investigates the effects of confidence transformation (CT) of the character classifier outputs in handwritten Chinese text recognition. The classifier outputs are transformed to confidence values in three confidence types, namely, sigmoid, soft max and Dempster-Shafer theory of evidence (D-S evidence). The confidence parameters are optimized by minimizing the cross-entropy (CE) loss function (both binary and multi-class) on a validation dataset, where we add non-character samples to enhance the outlier rejection capability in text recognition. Experimental results on the CASIA-HWDB database show that confidence transformation improves the handwritten text recognition performance significantly and adding non-characters for confidence parameter estimation is beneficial. Among the confidence types, the D-S evidence performs best.
AB - This paper investigates the effects of confidence transformation (CT) of the character classifier outputs in handwritten Chinese text recognition. The classifier outputs are transformed to confidence values in three confidence types, namely, sigmoid, soft max and Dempster-Shafer theory of evidence (D-S evidence). The confidence parameters are optimized by minimizing the cross-entropy (CE) loss function (both binary and multi-class) on a validation dataset, where we add non-character samples to enhance the outlier rejection capability in text recognition. Experimental results on the CASIA-HWDB database show that confidence transformation improves the handwritten text recognition performance significantly and adding non-characters for confidence parameter estimation is beneficial. Among the confidence types, the D-S evidence performs best.
KW - Handwritten text recognition
KW - confidence transformation
KW - cross-entropy
KW - non-characters
UR - http://www.scopus.com/inward/record.url?scp=82355168523&partnerID=8YFLogxK
U2 - 10.1109/ICDAR.2011.110
DO - 10.1109/ICDAR.2011.110
M3 - Conference Proceeding
AN - SCOPUS:82355168523
SN - 9780769545202
T3 - Proceedings of the International Conference on Document Analysis and Recognition, ICDAR
SP - 518
EP - 522
BT - Proceedings - 11th International Conference on Document Analysis and Recognition, ICDAR 2011
T2 - 11th International Conference on Document Analysis and Recognition, ICDAR 2011
Y2 - 18 September 2011 through 21 September 2011
ER -