ISOLARED HANDWRITTEN ARABIC NUMERALS RECOGNITION USING THE K- NEAREST NEIGHBOR AND THE HIDDEN MARKOV MODEL CLASSIFIERS

Abstract

This work deals with a recognition system of handwritten Arabic numerals
extracted to the MNIST standard database (Arabic numerals), this system is composed
by three main phases: the preprocessing of numerals followed by the extraction of primitives
with the zoning method in order to convert each image into a vector number which
is nothing other than an information extracted from this numeral just to differentiate
the others. Finally, our recognition system will end with a classification phase by the
two methods: the K-nearest neighbours (K-NN) and Hidden Markov Model (HMM).
This work has achieved a recognition rate of approximately 82 of success.