基于L*a*b*色彩空间的视神经边缘自动提取

视神经(Optic nerve)形状、面积和深度等参数是衡量眼底健康状况的重要指标, 其边缘提取是量化这些参数的前提. 为精确识别视神经边缘, 本文提出了一种基于L*a*b*色彩空间眼底图像视神经边缘自动提取算法. 该方法通过L*a*b*色彩空间自适应形态学方法与区域辅助几何活动轮廓模型边缘提取方法, 结合基于交叉网络的视神经自动定位, 实现视神经边缘的自动提取. 采用国际上通用的DRIVE眼底图像库和临床图像进行实验, 验证了该算法的有效性.

Automated Detection of Optic Nerve Boundary Based on L~*a~*b~* Color Space

The parameters of optic nerve, such as shape, area, depth and so on, are the important indices to estimate the retinal health. Extraction of the optic nerve boundary is the precondition for quantifying these parameters. To accurately identify the boundary, a new method to detect the exact optic nerve boundary is proposed based on the L*a*b* color space in this paper. Using morphology and region-aided geometric snake in the L*a*b* color space, the method applies the automatic optic nerve localization base on cross-network to realize the automatic extraction of optic nerve boundary. The feasibility of the new method has been successfully tested for the DRIVE fundus image databases and our clinic images.