Qualitative and quantitative analysis for facial complexion in traditional Chinese medicine

Abstract

Facial diagnosis is an important and very intuitive diagnostic method in Traditional Chinese Medicine (TCM). However, due to its qualitative and experience-based subjective property, traditional facial diagnosis has a certain limitation in clinical medicine. The computerized inspection method provides classification models to recognize facial complexion (including color and gloss). However, the previous works only study the classification problems of facial complexion, which is considered as qualitative analysis in our perspective. For quantitative analysis expectation, the severity or degree of facial complexion has not been reported yet. This paper aims to make both qualitative and quantitative analysis for facial complexion. We propose a novel feature representation of facial complexion from the whole face of patients. The features are established with four chromaticity bases splitting up by luminance distribution on CIELAB color space. Chromaticity bases are constructed from facial dominant color using two-level clustering; the optimal luminance distribution is simply implemented with experimental comparisons. The features are proved to be more distinctive than the previous facial complexion feature representation. Complexion recognition proceeds by training an SVM classifier with the optimal model parameters. In addition, further improved features are more developed by the weighted fusion of five local regions. Extensive experimental results show that the proposed features achieve highest facial color recognition performance with a total accuracy of 86.89%. And, furthermore, the proposed recognition framework could analyze both color and gloss degrees of facial complexion by learning a ranking function.

Figure 1

Pipeline of the proposed facial color classification framework.

Figure 2

Six typical facial skin regions after skin detection and fine-tuning.

Figure 3

Four chromaticity bases constructed by the proposed approach.

Figure 4

Six typical facial complexion histograms.

Figure 5

The confusion matrices produced by our model.

Figure 6

Comparisons of dominant color extraction methods on four facial colors.

Figure 7

Classification accuracy with different luminance distribution intervals.

Figure 8

An example of facial image segmentation.

Figure 9

The flowchart of local regions segmentation.

Figure 10

The confusion matrices produced by our improved model.

Figure 11

The ranking distribution according to facial color degree and gloss degree.