Tongue image color correction method based on root polynomial regression

Color correction is a necessary method step in the objectification of tongue diagnosis. Conventional polynomial regression is widely used in the online color correction, but it is easily influenced by the lighting source. In order to meet the requirements of different lighting conditions and maintain the accuracy of color correction, the Root-Polynomial Color Correction algorithm is used in this paper. This study established mapping between collected RGB value and standard RGB value through the calibration of the X-rite Color Checker. Finally, the CIELab color difference(ΔE) is calculated to verify the color correction effect. The color of the corrected images is close to the true color of the tongue. The average ΔE is less than 5, and the average algorithm calculation time t is less than 0.1 s. The experimental results show that RPCC can adapt to different lighting environments, and the color correction chromatic error is small.