In this paper, a histogram of oriented gradient (HOG)-assisted deep learning algorithm is proposed for classification of Manipuri Meitei Mayek (MMM) numerals. In HOG, gradient computes the magnitude and direction of largest change in intensity among small neighborhood of every pixel. The extracted HOG feature vectors are trained and recognized by using multilayer perceptron (MLP) artificial neural network. The stacking method is used to gain robust features. After that, it is applied to as Softmax classifier and aims for the recognition of MMM's character. To prove the effectiveness of our work, various comparisons are made on existing state of the art of Manipuri optical character recognition (OCR) algorithms and it shows that our proposed algorithm is superior comparing to the other existing Manipuri OCR work.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.