Character Analysis of Facial Expression Thermal Image

Abstract

Facial expression recognition will be studied in this paper using mathematics morphology, nose min-temperature and temperature histogram through drawing and analyzing the whole geometry characteristics and local characteristics of the interesting area of infrared thermal imaging (IRTI). The author has completed the automatic calculation of the area, girth, circularity and angles, which is supposed to be the parameter vector of interesting region. The results show that geometry characteristics in the interesting region of different expression are obviously different; Facial temperature changes almost with the expression at the same time. Studies have shown feasibility of facial expression recognition on the basis of IRTI. This method can be used to monitor the facial expression in real time, which can be used in auxiliary diagnosis and medical on disease.