摘要
OCT(光学相干层析技术)自从1991年出现以来,在多个领域展现出了应用前景,尤其是在眼科疾病的检测方面起了重要作用。视网膜的分层,厚度检测就是利用OCT技术所得图像,对眼底的各种疾病的预防和治疗给出依据。然而,由于散斑噪声,低对比度,不规则的形态结构等原因,使得视网膜的分层检测存在不小的难度。为了解决这些问题,提高电脑对OCT图像处理的能力,文中提出了一种分为两步的图像分割算法。第一步是通过标记的分水岭算法进行初次分割;第二步是利用不同的区域特征值进行最优化合并,以获取需要的边界信息。实验结果表明,对于平滑的部分,分层有相对较好的结果,面对病变区域的处理仍有可改进的部分。
OCT （optical coherence tomography） which appears in 1991 has showed the application prospects in a number of areas, in particular, played an important role in ophthalmology. With OCT images, segmentation of the retina and detection of the retinal thickness give the evidence for the prevention and treatment of various diseases of eyes. However, due to speckle noise, low contrast, irregular morphology, makes the retinal segmentation becomes difficult. In order to solve these problems and to improve the ability of the computer to the OCT image processing, this paper proposes a two-step image segmentation algorithm. The first step is through marked watershed algorithm to the initial segmentation ; the second step is to use different region characteristics to optimize merge in order to obtain the required boundary information. The experimental results show that the smooth portion has relatively good effect, but there is still room to improve in the processing of the lesion area portion.
出处
《信息技术》
2013年第4期136-140,共5页
Information Technology
