Pandey, Ram Chandra2022-01-302022-01-302016-10Department of Electronics and Computer Engineeringhttps://hdl.handle.net/20.500.14540/7811The OCR systems developed for the Nepali language carry a very poor recognition rate due to error in character segmentation, ambiguity with similar character, unique character representation style.The OCR systems developed for the Nepali language carry a very poor recognition rate due to error in character segmentation, ambiguity with similar character, unique character representation style. The purpose of this thesis work is to take image of handwritten or printed Nepali characters and words as input, process the character, train the neural network algorithm, to recognize the pattern and convert to digital form of the input. In this thesis, proposing an OCR for Nepali text in Devanagari script, using multi-layer feed forward back propagation Artificial Neural Network (ANN), which will improve its efficiency and accuracy. Adaptive learning rate with Gradient descent algorithm is proposed in Neural net with two hidden layers used with input and output and MMSE is the performance criteria. Various classifiers for training characters need to be created and stored.enNeural Network,Nepali Handwritten Datasets,Handwriting Recognition.Thesis