Wyniki wyszukiwania - BazTech

Ograniczanie wyników

1 Systems Science

1 2010

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: Human Computer Interaction System

Sortuj według:

Ogranicz wyniki do:

An adaptive heuristic model for optimizing the fuzzy expressions classification

Jamshidnezhad A., Nordin J. M.

Systems Science

2010

Vol. 36, no 4

83-91

Facial expression recognition is an advanced step for Human Computer Interaction (HCI) systems. Recently, fuzzy techniques are used widely to solve the natural based problems in which ambiguity is an inherent matter. In this paper, a Genetic Algorithm as a novel heuristic process is modeled to optimize the performance of fuzzy system to recognize facial expression from images. In the proposed hybrid model the core of expression recognition system is a Mamdani-type fuzzy rule based system to recognize the emotions; also, a proposed Genetic Algorithm is used with the purpose of making better performance and parameter optimization to improve the accuracy and robustness of the system. Therefore, GA as a training technique sets the fuzzy membership functions under the adverse conditions. To evaluate the system performance, images from FG-Net (FEED) and Cohn-Kanade database were used to obtain the best function parameters. Results showed the hybrid model under the training process not only to increase the accuracy rate of emotion recognition but also to increase the validity of the model in adverse conditions.