Using the way combining the mathematic morphology top - hat and gauss matched filtering can effectively reduce the influence of the inherent fovea and the marginal noise brought forth by the retinal camera caliber in retinal image and increase the precision of retinal image segmentation and feature point extraction 文中提出了采用灰度数学形态学中的top - hat变换和高斯匹配滤波相结合的方法，用来有效的降低视网膜荧光图像中固有的小窝现象及眼底照相机的口径所引起的视网膜图像边界像素噪声，极大的提高了图像分割及特征提取的精度。
