Devanagari digit recognition by using artificial neural network
Reena Dhakad, Dinesh Soni · 2017
This paper presents a brand new offline written syllabic script digit recognition system. although syllabic script is that the script for Hindi, that is that the official language of India, its character and digit recognition create nice challenges because of massive type of symbols and their proximity in look.so as to extract options which might distinguish similar showing digits, we have a tendency to use Neural network. The resultant massive dimensional feature house is handled by careful application of Principal element Analysis (PCA). The Support Vector Machine (SVM) and k-NN categoryifiers were used with one-against-rest class model. Results of Neural network feature extractor and classifiers have shown that NN with k-NN gave overall higher results than the Neural network classifier and shown highest results (93.21%) accuracy on a syllabic script written digits set.