Connan, JamesNaidoo, Nathan LyleDept. of Computer ScienceFaculty of Science2013-12-102024-10-302011/02/172011/02/172013-12-102024-10-302010https://hdl.handle.net/10566/16949Masters of ScienceThis 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%.enOptical pattern recognitionMathematical modelsImage processingDigital techniquesMarkov processesThesisUniversity of the Western Cape