Digital tongue image analysis in medical applications using a new tongue ColorChecker

摘要

Digital tongue images are analyzed by researchers for disease detection. In order to ensure a better color representation of the tongue body leading to more accurate detection results, this paper proposes a new tongue ColorChecker. This ColorChecker will act as a color reference, and can be used in color calibration algorithms to standardize the captured digital tongue images. Initially, a statistical tongue color gamut is established based on a large tongue image dataset. Through experimentation, various quantities of color-patches on the tongue ColorChecker are determined. Afterwards, the results produced by the Munsell ColorChecker are compared with the proposed tongue ColorChecker by applying the color difference calculation formula of CIELAB 1976 as a reference for the mean color calibration error. The results demonstrate that the proposed tongue ColorChecker, which has 24 color-patches, produces a smaller error (CIELAB 1976 - 10.6058) compared with the Munsell ColorChecker (CIELAB 1976 - 14.7836).