An MLP based Approach for Recognition of Handwritten `Bangla' Numerals

Subhadip Basu; Nibaran Das; Ram Sarkar; Mahantapas Kundu; Mita Nasipuri; Dipak Kumar Basu

An MLP based Approach for Recognition of Handwritten `Bangla' Numerals

Computer Vision and Pattern Recognition 2012-03-06 v1 Artificial Intelligence

Authors: Subhadip Basu , Nibaran Das , Ram Sarkar , Mahantapas Kundu , Mita Nasipuri , Dipak Kumar Basu

Abstract

The work presented here involves the design of a Multi Layer Perceptron (MLP) based pattern classifier for recognition of handwritten Bangla digits using a 76 element feature vector. Bangla is the second most popular script and language in the Indian subcontinent and the fifth most popular language in the world. The feature set developed for representing handwritten Bangla numerals here includes 24 shadow features, 16 centroid features and 36 longest-run features. On experimentation with a database of 6000 samples, the technique yields an average recognition rate of 96.67% evaluated after three-fold cross validation of results. It is useful for applications related to OCR of handwritten Bangla Digit and can also be extended to include OCR of handwritten characters of Bangla alphabet.

Keywords

handwritten digit recognition multi-layer perceptron feature extraction bangla script optical character recognition pattern classification

Cite

@article{arxiv.1203.0876,
  title  = {An MLP based Approach for Recognition of Handwritten `Bangla' Numerals},
  author = {Subhadip Basu and Nibaran Das and Ram Sarkar and Mahantapas Kundu and Mita Nasipuri and Dipak Kumar Basu},
  journal= {arXiv preprint arXiv:1203.0876},
  year   = {2012}
}

An MLP based Approach for Recognition of Handwritten `Bangla' Numerals

Abstract

Keywords

Cite

Related papers