South African sign language recognition using feature vectors and Hidden Markov Models
Nathan Lyle Naidoo · University of the Western Cape Electronic Theses and Dissertations Repository (University of the Western Cape) · 2010
This thesis presents a system for performing whole gesture recognition for South African Sign Language.The system uses feature vectors combined with Hidden Markov models.In order to constuct a feature vector, dynamic segmentation must occur to extract the signer's hand movements.Techniques and methods for normalising variations that occur when recording a signer performing a gesture, are investigated.The system has a classification rate of 69%.First and foremost I wish to thank the Lord Jesus Christ, the Son of God, who gave me the use of all my faculties and a sound mind, apart from which I would not be able to complete this thesis.I wish to thank my father, Prof. Anthony Naidoo and my late mother, Charmaine Naidoo for instilling in me the need for education.To my twin brother, Marc Naidoo who has stood by