Title :
Fuzzy Model Based Recognition of Handwritten Hindi Numerals using Bacterial Foraging
Author :
Hanmandlu, M. ; Nath, A.V. ; Mishra, A.C. ; Madasu, V.K.
Author_Institution :
Indian Inst. of Technol. Bombay, Mumbai
Abstract :
This paper presents the recognition of handwritten Hindi numerals. The recognition is based on the modified exponential membership function fitted to the fuzzy sets derived from features consisting of normalized distances obtained using the Box approach. The exponential membership function is modified by two structural parameters that are estimated by optimizing the entropy subject to the attainment of membership function to unity. The optimization strategy used is the foraging model of E.coli bacteria. Two window sizes are used: one for ( ) and another for the rest of the numerals. Experimentation is carried out on a limited database of nearly 3500 samples. The overall recognition is found to be 96%.
Keywords :
feature extraction; fuzzy set theory; handwritten character recognition; natural languages; optimisation; bacterial foraging; exponential membership function; feature extraction; fuzzy set; handwritten Hindi numeral recognition; optimization; Character recognition; Databases; Entropy; Feature extraction; Fuzzy sets; Handwriting recognition; Microorganisms; Neural networks; Parameter estimation; Structural engineering;
Conference_Titel :
Computer and Information Science, 2007. ICIS 2007. 6th IEEE/ACIS International Conference on
Conference_Location :
Melbourne, Qld.
Print_ISBN :
0-7695-2841-4
DOI :
10.1109/ICIS.2007.103