眼科生物参数测量仪自动对准及对焦方法  

Ophthalmic biological parameter measuring instrument automatic alignment and focusing method

作　　者：刘桐生 曾亚光 何志敏 LIU Tongsheng;ZENG Yaguang;HE Zhimin(School of Computing and Artificial Intelligence,Foshan University,Foshan 528225,China;School of Physics and Optoelectronic Engineering,Foshan University,Foshan 528225,China;School of Electronic and Information Engineering,Foshan University,Foshan 528225,China)

机构地区：[1]佛山大学计算机与人工智能学院,广东佛山528225 [2]佛山大学物理与光电工程学院,广东佛山528225 [3]佛山大学电子信息工程学院,广东佛山528225

出　　处：《佛山科学技术学院学报(自然科学版)》2025年第2期37-42,共6页Journal of Foshan University(Natural Science Edition)

基　　金：广东省自然科学基金资助项目(2022A1515140116)。

摘　　要：为了解决现有眼科生物参数测量仪对准对焦速度慢、精度低和普适性差的问题,提出了基于灰度能量曲线的光轴中心自动对准算法以及基于决策树的自动对焦算法。首先,对仪器实拍的人眼图像进行局部区域高斯模糊预处理,再利用行和列方向的灰度能量曲线定位靶环感兴趣区域。最后,经过图像后处理可以得到完整的靶环区域,用光斑质心拟合椭圆的方法来计算光轴中心位置。此外,还利用靶环区域作为先验知识,通过提取靶环大小和清晰度特征构建数据集,并训练决策树分类模型区分准焦-近离焦-远离焦图像,实现仪器自动对焦。实验结果表明,所提算法的定位误差在3像素以内,单帧图像的对准和对焦平均处理时间为14.8ms,具有实时性、高精度和鲁棒性强等优点,满足眼科生物参数测量自动对准和对焦的要求。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.

关 键 词：自动对准 自动对焦 中心定位 灰度能量曲线 决策树 图像处理 眼科医疗设备 

分 类 号：R770.4[医药卫生—眼科]