Spatial Similarity based Stroke Number and Order Free Clustering

Santosh K.C. (1) , Cholwich Nattee (2) , Bart Lamiroy (1, 3)

1 QGAR - Querying Graphics through Analysis and Recognition
2 KINDML - Knowledge Information & Data Management Laboratory
3 ORPAILLEUR - Knowledge representation, reasonning

In this paper, we present an innovative approach to integrate spatial relations in stroke clustering for handwritten Devanagari character recognition. It handles strokes of any number and order, writer independently. Learnt strokes are hierarchically agglomerated via Dynamic Time Warping based on their location and their number and stored accordingly. We experimentally validate our concept by showing its ability to improve recognition performance on previously published results.

Mots clés: Stroke Spatial Information; Stroke Number and Order free; Hierarchical Clustering; Devanagari Handwriting Recognition

Domaines: Vision par ordinateur et reconnaissance de formes [cs.CV]

Santosh K.C., Cholwich Nattee, Bart Lamiroy. Spatial Similarity based Stroke Number and Order Free Clustering. International Conference on Frontiers in Handwriting Recognition, Nov 2010, Kolkata, India. ⟨inria-00516726⟩