Abstract: Recognition of handwritten text has been one of the active and challenging areas of research in the field of image processing and pattern recognition. It has numerous applications which include postal mail application, reading aid for blind and conversion of any handwritten document into electronic form. In this paper we focus on recognition of Karnataka district names in Kannada and 20 different English words from a given scanned word image with the help of classifiers. The first step is image acquisition which acquires the scanned image followed by noise filtering, smoothing and resizing of scanned image. Feature Extraction improves recognition rate and misclassification. We use edge detection algorithm to extract sharp edges and features are extracted to train the classifier to classify and recognize the handwritten word.

Keywords: Handwritten word recognition, feature extraction, Dominant Points, Edge Detection algorithm and Classification (key words).