Title :
Automatic recognition of magnetic cheque characters with hidden Markov models
Author :
Strydom, D. ; du Preez, J.A. ; Mostert, S.
Author_Institution :
Dept. of Electr. & Electron. Eng., Univ. of Stellenbosch, South Africa
Abstract :
Given a signal with a distinct pattern, one can easily define a model that represents such a signal as a sequence of states each with unique features. The models are hidden Markov models (HMM) configured in the most obvious way dependent only on the segmentation technique suggested. Each segment represents a state in the HMM. Models are proposed for recognition purposes
Keywords :
character recognition; hidden Markov models; HMM; automatic character recognition; hidden Markov models; magnetic cheque characters; segmentation technique; Character recognition; Feature extraction; Feeds; Hidden Markov models; Magnetic flux; Mechanical variables measurement; Pattern recognition; Time measurement; Time varying systems; Velocity measurement;
Conference_Titel :
Communications and Signal Processing, 1993., Proceedings of the 1993 IEEE South African Symposium on
Conference_Location :
Jan Smuts Airport
Print_ISBN :
0-7803-1292-9
DOI :
10.1109/COMSIG.1993.365866
