一种用于白内障眼底图像的无标注修复网络
图像与视频处理
2022-10-19 v1 计算机视觉与模式识别
摘要
白内障是全球视力丧失的首要原因。人们开发修复算法以改善白内障眼底图像的可读性，从而提升白内障患者诊断与治疗的确定性。遗憾的是，对标注的需求限制了这些算法在临床中的应用。本文提出一种无需标注即可修复白内障眼底图像的网络（ArcNet），以提升修复的临床实用性。ArcNet 无需标注，其从眼底图像中提取高频分量以替代分割来保留视网膜结构。该修复模型从合成图像中学习并适配至真实白内障图像。我们进行了大量实验以验证 ArcNet 的性能与有效性。ArcNet 相较于最先进（SOTA）算法取得了优异性能，且促进了白内障患者眼底疾病的诊断。在缺乏标注数据情况下正确修复白内障图像的能力，使所提算法具备卓越的临床实用性。
引用
@article{arxiv.2203.07737,
title = {An Annotation-free Restoration Network for Cataractous Fundus Images},
author = {Heng Li and Haofeng Liu and Yan Hu and Huazhu Fu and Yitian Zhao and Hanpei Miao and Jiang Liu},
journal= {arXiv preprint arXiv:2203.07737},
year = {2022}
}
备注
Copyright 2022 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works