作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2020, Vol. 46 ›› Issue (3): 267-272,279. doi: 10.19678/j.issn.1000-3428.0054379

• 图形图像处理 • 上一篇    下一篇

基于密集注意力网络的视网膜血管图像分割

梅旭璋, 江红, 孙军   

  1. 华东师范大学 计算机科学与软件工程学院, 上海 200062
  • 收稿日期:2019-03-26 修回日期:2019-05-06 发布日期:2020-03-14
  • 作者简介:梅旭璋(1995-),男,硕士研究生,主研方向为计算机视觉;江红(通信作者),副教授、博士;孙军,硕士研究生。
  • 基金资助:
    数学工程与先进计算国家重点实验室开放基金(2016A05)。

Retinal Vessel Image Segmentation Based on Dense Attention Network

MEI Xuzhang, JIANG Hong, SUN Jun   

  1. School of Computer Science and Software Engineering, East China Normal University, Shanghai 200062, China
  • Received:2019-03-26 Revised:2019-05-06 Published:2020-03-14

摘要: 视网膜血管的结构信息对眼科疾病的诊断具有重要的指导意义,对视网膜血管图像进行高效正确的分割成为临床的迫切需求。传统的人工分割方法耗时较长且易受个人主观因素的影响,分割质量不高。为此,提出一种基于密集注意力网络的图像自动分割算法。将编码器-解码器全卷积神经网络的基础结构与密集连接网络相结合,以充分提取每一层的特征,在网络的解码器端引入注意力门模块,对不必要的特征进行抑制,提高视网膜血管图像的分割精度。在DRIVE和STARE眼底图像数据集上的实验结果表明,与其他基于深度学习的算法相比,该算法的敏感性、特异性、准确率和AUC值均较高,分割效果较好。

关键词: 图像分割, 视网膜血管, 全卷积神经网络, 密集连接, 注意力机制

Abstract: The structural information of retinal vesselsassists in the diagnosis of ophthalmic diseases,and thus efficient and accurate segmentation of retinal vessel images has become an urgent clinical demannd.The traditional artificial segmentation methods are time-consumingand frequently affected by personal subjective factors,leading to a decline in segmentation quality.To address the problem,thispaper proposes an automatic image segmentation algorithm based on dense attention network.The algorithm combines the basic structure of the encoder-decoder fully convolutional neural network with the densely connected network to fully extract the features of each layer.Then the attention gate module on the decoder side of the network is introduced to suppress unnecessary features and thus improve the segmentation accuracy of retinal vessel segmentation.Experimental results on DRIVE and STARE fundus image datasets show that compared with other algorithms based on deep learning,the proposedalgorithm has excellent segmentation performance with the sensitivity,specificity,accuracy and AUC value all improved.

Key words: image segmentation, retinal vessel, Full Convolutional Neural Network(FCNN), dense connection, attention mechanism

中图分类号: