Recognition of handwritten Bengali characters: a novel multistage approach

A multistage scheme for the recognition of handwritten Bengali characters is introduced. An analysis of the Bengali character set has been carried out to isolate specific high-level features that can help in forming smaller sub-groups within the character set. This analysis demonstrates how detection of these various high-level features might help formulate successful multistage OCR design. A multiple expert decision combination hierarchy has been exploited to achieve higher performance from the proposed multi-stage framework.

[1]  Stavros Paschalakis,et al.  Novel multiple-stage approach to the recognition of handwritten words extracted from British cheques , 2000 .

[2]  Michael C. Fairhurst,et al.  Performance comparisons in hierarchical architectures for memory network pattern classifiers , 1986, Pattern Recognit. Lett..

[3]  Ke Chen,et al.  Methods of Combining Multiple Classifiers with Different Features and Their Applications to Text-Independent Speaker Identification , 1997, Int. J. Pattern Recognit. Artif. Intell..

[4]  Ahmad Fuad Rezaur Rahman,et al.  Multi-prototype classification: improved modelling of the variability of handwritten data using statistical clustering algorithms , 1997 .

[5]  Fuad Rahman,et al.  Machine-printed character recognition revisited: re-application of recent advances in handwritten character recognition research , 1998, Image Vis. Comput..

[6]  Josef Kittler,et al.  Weighting Factors in Multiple Expert Fusion , 1997, BMVC.

[7]  Majid Ahmadi,et al.  Recognition of handwritten numerals with multiple feature and multistage classifier , 1995, Pattern Recognit..

[8]  Stavros Paschalakis,et al.  A novel multiple-expert approach to the recognition of handwritten words extracted from British cheques , 1999 .

[9]  C.Y. Suen,et al.  Analysis and recognition of alphanumeric handprints by parts , 1994, IEEE Trans. Syst. Man Cybern..

[10]  Ching Y. Suen,et al.  The behavior-knowledge space method for combination of multiple classifiers , 1993, Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[11]  Fuad Rahman,et al.  A new hybrid approach in combining multiple experts to recognise handwritten numerals , 1997, Pattern Recognit. Lett..

[12]  Ahmad Fuad Rezaur Rahman,et al.  Enhancing multiple expert decision combination strategies through exploitation of a priori information sources , 1999 .

[13]  Ching Y. Suen,et al.  Application of majority voting to pattern recognition: an analysis of its behavior and performance , 1997, IEEE Trans. Syst. Man Cybern. Part A.

[14]  Mohammad Kaykobad,et al.  A Complete Bengali OCR: A Novel Hybrid Approach to Handwritten Bengali Character Recognition , 1998 .

[15]  Geoffrey E. Hinton,et al.  Learning internal representations by error propagation , 1986 .

[16]  Ahmad Fuad Rezaur Rahman.  Revised connected component algorithm and similarity measure quotient; application in handwritten Bengali character recognition , 1998 .

[17]  Ahmad Fuad Rezaur Rahman,et al.  Exploiting second order information to design a novel multiple expert decision combination platform for pattern classification , 1997 .

[18]  Ahmad Fuad Rezaur Rahman,et al.  Generalised approach to the recognition of structurally similar handwritten characters using multiple expert classifiers , 1997 .

[19]  Fuad Rahman,et al.  An Evaluation Of Multi-Expert Configurations For The Recognition Of Handwritten Numerals , 1998, Pattern Recognit..

[20]  Sargur N. Srihari,et al.  Decision Combination in Multiple Classifier Systems , 1994, IEEE Trans. Pattern Anal. Mach. Intell..