Facial expression recognition based on ASM and Finite Automata Machine

According to the problem that the facial expression recognition algorithm is easy to be affected by the error of the feature detection, this paper proposes a new algorithm based on Finite Automata Machine and realize the complete facial expression recognition system. The implementation steps include image preprocessing, face detection based on Adaboost, facial features localization based on ASM, normalization and facial expression classification based on Finite Automata Machine. Then experimental results show that this new facial recognition system possesses high accuracy, high stability and fast running speed. This provides a new method for the improvement of facial expression recognition, which promotes the development of artificial intelligence and human-computer interaction as well.