Design and Simulation of Handwritten Text Recognition System

Abstract

In this paper, the proposed approach for handwriting recognition system preprocessing, segmentation and feature extraction with neural network for character recognition. Input is digitized image containing any text, which is preprocessed to segment it into normalized individual words. Further feature extraction is used for extracting the features of handwritten alphabets. A neural network is trained onto the dataset containing some samples for each of the alphabets for recognition. A new approach for character recognition is implemented in this paper which segments character recognition from the text, which improves the accuracy significantly. A neural network is being used for character classification which also helps in deciding the threshold value for the character separation from the running text word. The software tool used is Labview. The proposed recognition system performs excellently for printed text and separate character written documents with 99.9% accuracy and for cursive handwriting with 70-80% accuracy. And successfully recognized for single, double word and so on and also recognized for a complete sentence.