Offline Handwritten Character Recognition Using Neural Network

This paper is aimed at recognition of offline handwritten characters in a given scanned text document with the help of neural networks. Image preprocessing, segmentation and feature extraction are various phases involved in character recognition. The first step is image acquisition followed by noise filtering, smoothing and image normalization of scanned image. Segmentation decomposes image into sub images and feature extraction extracts features from input image. Neural Network is created and trained to classify and recognize handwritten characters.