Detection of the Pupil Eigenvalues in Medicine

In order to detect the pupil for medical needs, An algorithm was proposed to obtain the pupil eigenvalues, which was based on pupil geometry and gray features, and used mathematical morphology operations. Firstly, depending on the gray scale features of pupil, the rough center of pupil was determined by gray projection. Then, based on the rough center, the sub-image that contains the pupil was intercepted with a rectangular window. After that, according to the geometrical characteristics that the pupil approximats a circle, spots were removed by reconstruction with a disc-shaped structural element, Furthermore, the interference of eyelashes was eliminated by the morphological operations. Finally, the pupil eigenvalues were obtained after edge extraction and circle hough transformation. The results show that the algorithm can effectively obtain the pupil eigenvalues, and that it is an objective, accurate, quantitative method to detect the pupil, so it provides physicians with reliable diagnostic clues.