Handwritten Character Recognition Using Convolutional Neural Network

Abstract: Handwritten character recognition has been one of the active and challenging research areas in the field of image processing and pattern recognition. It has numerous advantages such as reading aid for bank cheques, recognizing character from form applications etc. An attempt is made to recognize handwritten characters for English alphabets using CNN. FKI dataset which consists of English alphabets are made use of to train the neural network. FKI balanced dataset consist of 131,600 images of characters and 47 classes. The feature extraction technique is obtained by normalizing the pixel values. Pixel values will range from 0 to 255 which represents the intensity of each pixel in the image and they are normalized to represent value between 0 and 1. Convolutional neural network is used as a classifier which trains the FKI dataset. The work is extended by adding some more dataset to FKI dataset of characters from English language and training the model. The prediction for the given input image is obtained from the trained classifier.

Keywords: CNN, Handwritten Characters, Feature extraction