General text line extraction approach based on locally orientation estimation

Nazih Ouwayed (1) , Abdel Belaïd (1) , François Auger (2)

1 READ - READ
2 IREENA - Institut de Recherche en Energie Electrique de Nantes Atlantique EA4642

This paper presents a novel approach for the multi-oriented text line extraction from historical handwritten Arabic documents. Because of the multi-orientation of lines and their dispersion in the page, we use an image paving allowing us to progressively and locally determine the lines. The paving is initialized with a small window and then its size is corrected by extension until enough lines and connected components were found. We use the Snake for line extraction. Once the paving is established, the orientation is determined using the Wigner-Ville distribution on the histogram projection prole. This local orientation is then enlarged to limit the orientation in the neighborhood. Afterwards, the text lines are extracted locally in each zone basing on the follow-up of the baselines and the proximity of connected components. Finally, the connected components that overlap and touch in adjacent lines are separated. The morphology analysis of the terminal letters of Arabic words is here considered. The proposed approach has been experimented on 100 documents reaching an accuracy of about 98.6%.

Mots clés: Handwritten Arabic documents; text line segmentation; skew angle estimation; Snake; Wigner-Ville distribution; overlapping and touching lines.; overlapping and touching lines

Domaines: Traitement du texte et du document

Nazih Ouwayed, Abdel Belaïd, François Auger. General text line extraction approach based on locally orientation estimation. Document Recognition and Retrieval XVII - DRR 2010, 17th Document Recognition and Retrieval Conference, Jan 2010, San Jose, CA, United States. pp.1-10. ⟨hal-00488359⟩