Optical Character Recognition System for Urdu Words in Nastaliq Font

Optical Character Recognition (OCR) has been an attractive research area for the last three decades and mature OCR systems reporting near to 100% recognition rates are available for many scripts/languages today. Despite these developments, research on recognition of text in many languages is still in its early days, Urdu being one of them. The limited existing literature on Urdu OCR is either limited to isolated characters or considers limited vocabularies in fixed font sizes. This research presents a segmentation free and size invariant technique for recognition of Urdu words in Nastaliq font using ligatures as units of recognition. Ligatures, separated into primary ligatures and diacritics, are recognized using right-to-left HMMs. Diacritics are then associated with the main body using position information and the resulting ligatures are validated using a dictionary. The system evaluated on Urdu words realized promising recognition rates at ligature and word levels. Keywords—Optical Character Recognition; Urdu Text; Ligatures; Hidden Markov Models; Clustering