用于视网膜血管分割的半监督深度学习框架北大核心CSCD

针对目前视网膜血管分割任务中伪标签质量参差不齐,获得高质量的伪标签需要经过筛选的问题,本文提出了一种新的用于视网膜血管分割的半监督深度学习框架。该框架采用分而治之的思想来处理数据,针对有标签数据,采用传统的深度学习方法;针对无标签数据,采用Mean teacher模型,通过对比同一输入的不同形态输出,让模型学习无标签数据之间的共同特征,避免了采用伪标签技术带来的筛选过程。本文将U型网络(u-neural networks,U-Net)、Dense-Net和Ladder-Net 3个基准网络放入该框架,在DRIVE和CHASEDB1数据集上进行训练测试,均取得了较好的分割效果,表明本文框架具有提高网络区分不同阈值像素的能力。

Semi-supervised deep learning framework for retinal vessel segmentation

In view of the problem that quality of pseudo-labels is uneven in the c urrent retinal vessel segmentation task and it requires to be screened to obtain the high-quality pseudo-labels,a novel semi-supervised deep learning framew ork for retinal vessel segmentation is proposed in this paper.The framework adopts the idea of divide and conquer to process data.Traditional deep learning metho ds are utilized especially for the labeled data,while Mean teacher model is use d to deal with the unlabeled data.By comparing the different morphological outp uts of the same input,the model can learn the common features between the unlab eled data and avoid the screening process brought by pseudo-label technology.T h ree benchmark networks,u-neural networks (U-Net),Dense-Net and Ladder-Net are put into the fra mew ork,the experiments are carried out on DRIVE and CHASEDB1 datasets,which achi eve good segmentation results.It shows that the framework can improve the abil ity of the network to distinguish different threshold pixels.