Abstract

Off-line handwriting recognition involves the automatic conversion of text in an image into letter codes which are usable for text-processing and other related applications. It is a quite difficult task as the writing styles of different individuals vary person to person. It has been an active area of research and due to its diverse applicable environment it continues to be a challenging research topic. Slant is the clockwise or anticlockwise angle between the vertical direction and the vertical text strokes. It is one of the most obvious measurable factors of different handwriting styles. In handwritten word recognition, slant estimation and correction is a widely used preprocessing technique to improve segmentation performance, which significantly affects recognition accuracy. Various slant estimation and correction techniques have been proposed till date. But it is found that most of techniques are not capable of 100% slant estimation. In this paper some of the slant estimation and correction techniques available in literature are discussed. It is a part of a research work for development of an efficient slant estimation and correction technique.

Links and resources

Tags