一种角膜Placido圆环边缘检测方法  

An Improved Pretreatment Algorithm of Corneal Topography

作　　者：郭雁文[1] 

机构地区：[1]中北大学光电仪器厂,山西太原030051

出　　处：《中北大学学报（自然科学版）》2014年第5期599-604,共6页Journal of North University of China(Natural Science Edition)

基　　金：山西省科技攻关项目(工业)(20120321028-01);山西省研究生创新重点项目(20113101)

摘　　要：基于Placido环的角膜地形图原始图像由于受到光斑、睫毛等不同噪声的影响,给图像的预处理造成了很大的困难.提出了一种基于极坐标系的角膜地形图像自适应预处理算法,重点研究了在极坐标情况下角膜地形图的自适应图像平滑处理,利用有用圆环在极坐标下呈现准直线规则,而睫毛等噪声信号以纵向线条或不同角度存在的特点,对平滑处理后的极坐标图像进行了自适应水平边缘检测,并对损失数据进行了有效的边缘生长,获得了角膜Placido圆环边缘检测完整的结果图像.实际检测数据表明:该算法具有数据损失小、精度高,算法易实现的特点,满足检测过程的实际使用要求.Based on Placido circle, original images of corneal topography were easily suffered from noise dis- turbances, such as faculae, eyelashes, resulting in difficulty in image pretreatment. A self-adaption pretreat- ment algorithm of corneal topography based on the polar coordinate system was presented. Self-adaption image smoothing processing of corneal topography was researched within the polar coordinate system. Self-adaption horizontal edge detection of the polar coordinate image which has been smoothed was carried out according to the principle of useful circles presenting approximate horizontal lines while noise signals presenting vertical lines or different angles within the polar coordinate system. Effective edge growth of lost data was achieved to get a complete image of Placido circle edge detection. Experimental data show that the algorithm is more suit- able for practical application, with the characteristics of less data loss, higher data accuracy and easier avail- ability.

关 键 词：角膜地形图 边缘检测 边缘生长 Placido圆环 图像预处理 

分 类 号：R772[医药卫生—眼科] TP391.41[医药卫生—临床医学]