Recognition on handwritten digits based on their topological and morphological properties
This paper presents a system for off-line handwritten numeral recognition based on topological properties of the digits. The first step in recognition algorithm is graph evaluation, obtained from the RLC of the digit, the second is measuring the geometrical properties of the elements of the graph and classification based on that measures. Algorithm was trained and tested on CEDAR database and it achieved correct recognition rate of 99,74%.
KeywordsFeature Vector Machine Intelligence Isomorphic Graph Handwritten Digit Neighbor Segment
- 1.P. G. Perotto, “A new method for automatic character recognition,” IEEE Transactions On Electronic Computers, vol. EC-12, pp. 521–526, Oct. 1963.Google Scholar
- 2.F. Ali and T. Pavlidis, “Syntactic recognition of handwritten numerals,” IEEE Transactions On System, Man, and Cybernetics, vol SMC-7, pp. 337–541, July 1977.Google Scholar
- 3.T. Pavlidis, “A vectorizer and feature extractor for document recognition,” Computer Vision, Graphics, and image processing, vol. 35, pp. 111–127, 1986.Google Scholar