针对视网膜血管分割中存在的细小血管像素模糊以及血管断裂的问题,本文提出一种改进的密集U型网络(dense residual U-shaped network,DRU-Net).首先,结合残差结构和密集连接的优点提出了密集残差模块,并用其构建DRU-Net网络的编码层和解码层,充分提取目标特征;然后在网络底部添加由空洞卷积搭建的多路特征蒸馏模块(multi-characteristic distillation block,MC-DB),提取不同尺度的图像特征信息;最后在网络的跳跃连接处引入双向卷积长短期记忆模块(bidirectional convolutional long and short-term memory,BConv LSTM),充分融合浅层和深层的特征信息,输出完整的血管图.在公开的数据集DRIVE和CHASE_DB1上进行实验,分别取得了0.966 9和0.976 4的准确度,同时AUC(area under curve)分别达到了0.983 9和0.986 7,证明网络具有较好的分割效果,拥有一定的应用价值.
Abstract
This paper proposes an improved dense residual U-shaped network(DRU-Net)to address the issues of blurry small vessel pixels and vessel discontinuity in retinal vessel segmentation.Firstly,the dense residual block(DRB)is proposed by combining the advantages of residual structure and dense connection,which is used to construct the encoding and decoding layers of the DRU-Net to fully extract the target features.Then,a multi-characteristic distillation module(MCDB)is added to the bottom of the network,which is built with dilated convolutions to extract image features at different scales.Finally,a bidirectional convolutional long short-term memory module(BConv LSTM)is introduced at the skip connection to fully fuse the shallow and deep features,and output the complete vessel map.Experimental results on the public datasets DRIVE and CHASE_DB1 achieve an accuracy of 0.966 9 and 0.9764,re-spectively.Meanwhile,the area under curve(AUC)reaches 0.9839 and 0.9867,respectively,which dem-onstrates that the network has good segmentation performance and certain application value.
维普
万方数据