Method for classifying bizygomatic colors of face on basis of support vector machine

The invention discloses a method for classifying bizygomatic colors of a face on the basis of a support vector machine, which comprises the following steps of: (1) establishing a bizygomatic color training sample set; (2) training the bizygomatic color training sample set by utilizing the support vector machine and constructing a bizygomatic color classifier; and (3) identifying by utilizing the bizygomatic color classifier and judging whether a bizygomatic image to be identified belongs to a red bizygomatic image or a non-red bizygomatic image or not. In the method, the support vector machine is used as the classifier, R, G and B values of pixels are used as feature vectors, a computer is utilized to simulate a Chinese medical specialist to identify the bizygomatic colors, and the methodhas repeatability; in the method, due to the adoption of a threshold value judging method in the identification phase, a bizygomatic color judging mechanism is simpler, more visual and more impartial; the Chinese medical spec