Abstract
Developing an automatic arabic sign language recognition system is of great importance, it can be used as a communication means between hearing-impaired and other people.
Such systems, are generally composed of two main stages : Hand detection and hand gesture recognition. To ensure these two steps, two versions of wavelet network classifiers will be used aiming at comparing their performances to employ the best one in our application.
These two classification engines are the wavelet network classifier learnt by fast wavelet transform (FWNC) and the separator wavelet network classifier (SWNC). The experimental results show the effectiveness of our proposed approaches of hand detection and hand gestures recognition.