Annotated dataset for retinal fundus image segmentation: Vascular arcade and optic nerve head

This repository contains a set of manually generated segmentation masks derived from retinal fundus images from the public APTOS 2019 Blindness Detection dataset. The annotations correspond to two relevant anatomical structures in ophthalmology: the vascular arcade and the optic nerve head. The dataset includes 500 masks in PNG format, along with annotation files in COCO (JSON) format and Python scripts used to generate the masks. This resource is intended to support the training and comparative evaluation of medical image segmentation models.本仓库包含一组从公开APTOS 2019失明检测数据集（APTOS 2019 Blindness Detection Dataset）的视网膜眼底图像中手动生成的分割掩码。本次标注对应眼科领域两类关键解剖结构：血管弓（vascular arcade）与视神经乳头（optic nerve head）。本数据集包含500份PNG格式的掩码文件，同时附带COCO（JSON）格式的标注文件与用于生成掩码的Python脚本。本资源旨在支撑医学图像分割模型的训练与对比评估工作。