Eyes-like diagnosis was a traditional Chinese medicine method for many diseases, such as chronic gastritis, diabetes, hypertension etc. There was a close relationship between viscera and eyes-like. White-Eye was divided into fourteen sections, which corresponded to different viscera, so eyes-like was the reflection of status of viscera, in another words, it was an epitome of viscera health condition. In this paper, we developed a novel shadowless imaging technology and system for eyes-like diagnosis in vivo, which consisted of an optical shadowless imaging device for capturing and saving images of patients’ eyes-like, and a computer linked to the device for image processing. A character matching algorithm was developed to extract the character of white-eye in corresponding sections of eyes-like images taken by the optical shadowless imaging device, according to the character of eyes-like, whether there were viscera diseases could be learned. A series of assays were carried out, and the results verified the feasibility of eyes-like diagnosis technique.