改进的密集U型网络对视网膜血管的分割  

Segmentation of retinal vessels based on a modified dense U-shaped network

作　　者：张晨 高文根 陈亮 李鹏飞 ZHANG Chen;GAO Wengen;CHEN Liang;LI Pengfei(School of Electrical Engineering,Anhui Polytechnic University,Wuhu,Anhui 241000,China;Key Laboratory of Advanced Perception and Intelligent Control for High-end Equipment,Ministry of Education,Wuhu,Anhui 241000,China)

机构地区：[1]安徽工程大学电气工程学院,安徽芜湖241000 [2]高端装备先进感知与智能控制教育部重点实验室,安徽芜湖241000

出　　处：《光电子．激光》2024年第4期414-422,共9页Journal of Optoelectronics·Laser

基　　金：国家自然科学基金(61976005)资助项目。

摘　　要：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,MCDB),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0.9669和0.9764的准确度,同时AUC(area under curve)分别达到了0.9839和0.9867,证明网络具有较好的分割效果,拥有一定的应用价值。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 0.976 4,respectively.Meanwhile, the area under curve(AUC) reaches 0.983 9 and 0.986 7,respectively, which demonstrates that the network has good segmentation performance and certain application value.

关 键 词：视网膜血管分割 密集残差模块(DRB) 多路特征蒸馏模块(MCDB) 双向卷积长短期记忆模块(BConv LSTM) 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]