M.S. Khorsheed (UK)
Arabic OCR, Handwriting, HMM, Viterbi Algorithm
An HMM is implemented to recognise handwritten Arabic words. The global model is composed of multiple left-to right HMMs where each model is trained using samples representing one letter from the alphabet. The skeleton of the words are transformed into a sequence of small line segments. The performance of the proposed method is as sessed using samples extracted from a historical handwirt ten manuscript.
Important Links:
Go Back
