Dccn: A Deep-Color Correction Network For Traditional Chinese Medicine Tongue Images

摘要

In Traditional Chinese Medicine (TCM), tongue inspection is one of the most important diagnostic means. Due to the limitations of capturing devices and variations in lighting conditions, there are color distortions between the captured tongue images and the actual human visual perceived. In this paper, we proposed a Deep Color Correction Network (DCCN) to learn the mapping model between the captured distorted color images and the target visually perceived color appearance under different lighting conditions and provides the color consistency across different cameras or capture deceives. Experimental results show that the DCCN model can achieve high accuracy and robustness in both of the objective and subjective tongue image color correction metrics.