论文部分内容阅读
眼底图像的快速自动拼接是眼底血管图像定量分析的必要前提。首先运用数学形态学的‘开’运算从原始图像中分离出灰度分布高度不均的背景图像。经图像增强，采用相位相关法求出图像间的平移量。结果表明：本文方法具有错配率低、速度快、拼接处光滑过渡并能同时增强血管对比度的优点
Fundus rapid automatic image stitching is a necessary prerequisite for quantitative analysis of fundus images. First of all, using the mathematical morphology of the “open” operation from the original image is separated from the uneven distribution of gray-scale background image. After the image is enhanced, the amount of translation between the images is calculated using the phase correlation method. The results show that the proposed method has the advantages of low mismatch rate, high speed, smooth transition at the splicing place and the simultaneous enhancement of blood vessel contrast