Canonical syllable segmentation of Telugu document images

Abstract

We propose a classical approach in the segmentation of canonical syllables of Telugu document images. Connected component approach is a simple strategy adopted until recent period. In this paper we propose a canonical syllable model which is equivalent to canonical structure C(C(CV))). The model consists of zone separation and component extraction phases as independent parts. The relation between zones and components is established in the segmentation process of canonical syllable. The segmentation efficiency of the proposed model is evaluated with respect to the canonical groups.