Diopter measurement method based on retina image processing

In order to solve the problem of border blur between dark background and bright ring image, a geometrical plane parameter measurement method based on retina images processing has been proposed in this paper. The key technique is to locate the image edge points by Gauss curve fitting and curvature filtering. Firstly, annular reflecting images are obtained by auto refractometer measurement system. Secondly, the coordinates of sub-pixel edge points are acquired through Gauss curve fitting method. Thirdly by curvature filtering method the noisy points are eliminated. Finally, the geometrical plane parameters of retina image are calculated after fitting ellipse track with the least-squares method. Thus human eye diopter can be calculated. The experimental results demonstrate the accuracy, stability and validity of the retina image processing method.