Invariant Moments Approach for Gujarati Numerals

Dr. Mamta Baheti

Due to less reported work for Gujarati numerals we have been motivated for same as Gujarati is a language not only of Indian states but widely spoken across world. We have. We have used noisy numerals for training and testing. Images are pre-processed and then subjected to the proposed algorithm. in our proposed algorithm we have used invariant moments as feature extraction technique and Gaussian distribution function as classifier. We found satisfactory results for some numerals. The results can be improved by giving better quality images for training and testing.