Recognition of human face using fuzzy inference and neural network
Toshio Fukuda, S. Itou, Fumihito Arai · 2003
This paper proposes an image extraction technique of human face by color image processing, fuzzy inference, and a neural network. The image processing technique provides a method of extraction of the skin color and measurement of the detailed information of a human face. The fuzzy inference estimates the facial direction based on the configuration of the extracted face candidates, while the neural network distinguishes faces from the others features using the detailed information and the facial, direction. This method makes it possible to recognize not only a full face but also side faces. Further, the authors show how this method works for the illusion problem. The extraction accuracy of this present system is 80 percent. According to the experimental results, this system can be used for the vision system of autonomous robots under a robot-human mixed environment.>