Segmentation of Cursive Handwritten Words using Hypergraph

Segmentation is a crucial step in the cursive script recognition process. A higher recognition rate is achieved if the characters of a word are correctly isolated. In this paper segmentation using Hypergraph model is investigated. Hypergraph model treats an image as packets of pixels. By recombining these packets of different sizes a given image can be segmented with the guarantee that at least one of the combinations provides a correct segmentation.