摘要
为医学上检测眼睛的需要,提出一种基于瞳孔灰度和几何特征运用数学形态学运算检测瞳孔特征值的算法。先根据瞳孔的灰度特点,用灰度投影方法确定出瞳孔的粗略中心点,以此为中心取一矩形窗口截取包含瞳孔的子图像。根据瞳孔近似圆形的几何特点,用圆盘形结构元素做重构开运算,消除光斑,运用形态学算子消除眼睫毛的干扰。通过边缘提取和圆的hough变换检测出瞳孔的边界点,获得瞳孔的特征值。试验结果表明,该算法能够有效地检测出红外和可见光眼睛图像中的瞳孔特征值,综合考虑了检测的精度和速度,满足医用眼睛图像检测的需要。
In order to detect the eyes for medical needs,an algorithm is proposed to detect the pupil eigenvalues using mathematical morphology operations based on pupil geometry and gray feature.Firstly,according to the gray scale features of the pupil,the rough center of the pupil is determined with gray projection method.Based on the rough center,the sub-image that contains the pupil is intercepted with a rectangular window.Then under the geometrical characteristics of the pupil that approximats circle,remove spots by reconstruction with disc-shaped structural elements.Further eliminate the interference of eyelash by the morphological operation.Finally, the eigenvalues of the pupil are obtained detecting the pupil boundary points by edge extraction and circle hough transform.The results show that the algorithmcan effectively detect the eigenvalues of the pupil,and the detection accuracy and speed meet the needs ofmedical eye image detection.
出处
《计算机工程与设计》
CSCD
北大核心
2010年第23期5124-5127,共4页
Computer Engineering and Design
