Abstract

Abstract: Devanagari Handwritten Character Recognition is the system which provides user the Optical CharacterRecognition functions on Devanagari handwriting. This Devanagari Handwritten character recognition system has mainly five stages: Preprocessing, Segmentation, Feature Extraction, Prediction and Postprocessing. It uses the Convolutional Neural Network to train the model and uses Image Processing techniques to recognize user image via camera or local storage. The system recommends the external camera with good image quality to get best results. Devanagari Handwritten Character Recognition system was carried out by the team of members, who were devoted to their respective field to develop the best possible. This system uses python programming language and different hardware platforms, software platforms to make the system as a whole. During the development phase different modules were made separately by different members and were integrated at last. Along with it testing were done. After the system has been developed it is found that different algorithms gave different level of correctness.Keywords Handwritten Devanagri alphabets; Handwritten Devanagri digits; Convolutional Neural Network; Segmentation; Templates matching