An automated feature extraction method for recognition of Petechia Spot in Tongue Diagnosis

Petechia Spot ranks itself among the most significant pathological features utilized in the practice of the Tongue Diagnosis because it can reveal the existence of tiny hemorrhage in some organ inside the body. In this paper, an automated feature extraction method for recognition of Petechia Spot in tongue image is proposed employing marker-based watershed transform and morphological filter. An efficient preprocessing method including gray-level contrast-stretch is designed before extraction to make Petechia Spot more apparent. Moreover, algorithm erasing falsely selected region is presented to avoid misjudgment. Experiment to test availability of the method is carried out, and result is proved to be satisfactory.