Off-line Telugu handwritten characters recognition using optical character recognition

Abstract

The Aim of the proposed paper is to recognize offline Hand written Telugu characters using Optical character recognition, OCR is one of the most popular and challenging topic of pattern recognition This paper proposes an OCR system for Telugu documents which comprises of three stages, namely pre-processing, feature extraction, and classification. In the preprocessing stage, we have employed median filtering on the input characters and applied normalization and skeletonization method over characters for extraction of boundary edge pixel points. In the feature extraction stage, initially the each character is divided into 3×3 grids and the corresponding centroid for all the nine zones are evaluated. With this we can identify the characters of different styles. There after, we have drawn the horizontal and vertical symmetric projection angel to the nearest pixel of the character which is dubbed as Binary External Symmetry Axis Constellation for unconstrained handwritten character. From which we have calculated the horizontal and vertical Euclidean distance for the same nearest pixel from centroid of each zone. Then we have calculated the mean Euclidean distance as well as the mean angular values of the zones. This is considered as the key feature values of our proposed system. Lastly, both support vector machine (SVM) and Quadratic discriminate Classifier (QDA) has been separately used as the classifier.