Ophthalmic biological parameter measuring instrument automatic alignment and focusing method

LIU Tongsheng; ZENG Yaguang; HE Zhimin

Abstract：In order to solve the problems of slow focusing speed,low accuracy and poor universality of the existing ophthalmic biological parameter measuring instruments,an automatic alignment algorithm of optical axis center based on grayscale energy curve and an automatic focus algorithm based on decision tree were proposed.Firstly,the local Gaussian blur preprocessing was carried out on the actual human eye image taken by the instrument,and then the grayscale energy curve in the row and column directions was used to locate the target ring of interest.Finally,the complete target ring region can be obtained after image post-processing,and the center position of the optical axis is calculated by fitting the ellipse with the center of mass of the spot.In addition,this study also used the target ring region as a priori knowledge,constructed a dataset by extracting the target ring size and clarity features,and trained a decision tree classification model to distinguish in-focus,near-defocus,and far-defocus images,so as to realize instrument autofocus.The experimental results show that the positioning error of the proposed algorithm is within 3 pixels,and the average processing time of alignment and focusing of a single frame image is 14.8ms,which has the advantages of real-time,high precision and robustness,which meets the requirements of automatic alignment and focusing of ophthalmic biological parameter measurement,and provides strong support for the development of intelligent ophthalmic medical equipment.