A complete system for Bangla handwritten numeral recognition

被引:25
作者
Pal, U.
Chaudhuri, B. B.
Belaid, A.
机构
[1] Indian Stat Inst, Comp Vis & Patten Recognition Unit, Kolkata 700108, India
[2] LORIA, Grp READ, F-54506 Vandoeuvre Les Nancy, France
关键词
handwritten segmentation; handwritten recognition; bangla numeral; Indian script;
D O I
10.1080/03772063.2006.11416437
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
This paper deals with a complete recognition system for unconstrained off-line Bangla (Bengali) handwritten numerals. To take care of variability involved in the writing style of different individuals, a robust scheme is presented here. The scheme is mainly based on new features obtained from the concept of water overflow from reservoir as well as topological and structural features of the numerals. The proposed scheme is tested on data collected from different individuals of various background and we obtained an overall recognition accuracy of about 92.8% from 12000 data.
引用
收藏
页码:27 / 34
页数:8
相关论文
共 16 条
[1]   Devnagari numeral recognition by combining decision of multiple connectionist classifiers [J].
Bajaj, R ;
Dey, L ;
Chaudhury, S .
SADHANA-ACADEMY PROCEEDINGS IN ENGINEERING SCIENCES, 2002, 27 (1) :59-72
[2]  
BHATTACHARYA, 2001, P ANN CONV COMP SOC, pC92
[3]   Integration of structural and statistical information for unconstrained handwritten numeral recognition [J].
Cai, JH ;
Liu, ZQ .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1999, 21 (03) :263-270
[4]   A complete printed Bangla OCR system [J].
Chaudhuri, BB ;
Pal, U .
PATTERN RECOGNITION, 1998, 31 (05) :531-549
[5]  
Chen YK, 2000, INT C PATT RECOG, P598, DOI 10.1109/ICPR.2000.906145
[6]   HANDWRITTEN NUMERAL RECOGNITION USING SELF-ORGANIZING MAPS AND FUZZY RULES [J].
CHI, ZR ;
WU, J ;
YAN, H .
PATTERN RECOGNITION, 1995, 28 (01) :59-66
[7]  
Congedo G., 1995, Proceedings of the Third International Conference on Document Analysis and Recognition, P1038, DOI 10.1109/ICDAR.1995.602080
[8]   BENGALI ALPHA-NUMERIC CHARACTER-RECOGNITION USING CURVATURE FEATURES [J].
DUTTA, A ;
CHAUDHURY, S .
PATTERN RECOGNITION, 1993, 26 (12) :1757-1770
[9]   A handwritten numeral character classification using tolerant rough set [J].
Kim, D ;
Bang, SY .
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2000, 22 (09) :923-937
[10]  
Kim KK, 2000, INT C PATT RECOG, P594, DOI 10.1109/ICPR.2000.906144