05
automatic optic disk and cup segmentation of fundus images using deep learning
2018
Method :分割,视盘视杯,杯盘比 ; one-stage
Dataset :Drishti-GS (160 training/10 validation / 51 testing)
Architecture : FCN
Results :Fscore
本文对于视杯视盘分割的related work 写的挺好挺全
Methods
眼底图像中视盘(OD)和杯状区域的自动分割 (CDR至少为0.65被认为是青光眼)
研究了CNN利用多个医生注释,对不同区域的像素进行优先级排序的各种策略。
-
Dataset
- Augment
水平垂直翻转,40 training set → 160 - center crop of 1200 × 1600
- training 160
- validation 10
- test 51
- Augment
-
FCN
-
stride = 8
-
数据量小(101),所以在 ImageNet 上 pre
-