MD-Net: A multi-scale dense network for retinal vessel segmentation

2021 ◽  
Vol 70 ◽  
pp. 102977
Author(s):  
Zhengjin Shi ◽  
Tianyu Wang ◽  
Zheng Huang ◽  
Feng Xie ◽  
Zihong Liu ◽  
...  
Keyword(s):  
Retinal Vessel ◽  
Vessel Segmentation ◽  
Dense Network ◽  
Multi Scale ◽  
Retinal Vessel Segmentation

Multi-Scale Retinal Vessel Segmentation Based on Fully Convolutional Neural Network

2019 ◽  
Vol 39 (2) ◽  
pp. 0211002 ◽  
Author(s):  
郑婷月 Zheng Tingyue ◽  
唐晨 Tang Chen ◽  
雷振坤 Lei Zhenkun
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Retinal Vessel ◽  
Vessel Segmentation ◽  
Multi Scale ◽  
Retinal Vessel Segmentation

scholarly journals Bi-SANet—Bilateral Network with Scale Attention for Retinal Vessel Segmentation

Symmetry ◽  
2021 ◽  
Vol 13 (10) ◽  
pp. 1820
Author(s):  
Yun Jiang ◽  
Huixia Yao ◽  
Zeqi Ma ◽  
Jingyao Zhang
Keyword(s):  
Spatial Information ◽  
Feature Fusion ◽  
Receptive Fields ◽  
Retinal Vessel ◽  
Vessel Segmentation ◽  
Low Contrast ◽  
Noise Interference ◽  
Multi Scale ◽  
Retinal Vessel Segmentation ◽  
Residual Block

The segmentation of retinal vessels is critical for the diagnosis of some fundus diseases. Retinal vessel segmentation requires abundant spatial information and receptive fields with different sizes while existing methods usually sacrifice spatial resolution to achieve real-time reasoning speed, resulting in inadequate vessel segmentation of low-contrast regions and weak anti-noise interference ability. The asymmetry of capillaries in fundus images also increases the difficulty of segmentation. In this paper, we proposed a two-branch network based on multi-scale attention to alleviate the above problem. First, a coarse network with multi-scale U-Net as the backbone is designed to capture more semantic information and to generate high-resolution features. A multi-scale attention module is used to obtain enough receptive fields. The other branch is a fine network, which uses the residual block of a small convolution kernel to make up for the deficiency of spatial information. Finally, we use the feature fusion module to aggregate the information of the coarse and fine networks. The experiments were performed on the DRIVE, CHASE, and STARE datasets. Respectively, the accuracy reached 96.93%, 97.58%, and 97.70%. The specificity reached 97.72%, 98.52%, and 98.94%. The F-measure reached 83.82%, 81.39%, and 84.36%. Experimental results show that compared with some state-of-art methods such as Sine-Net, SA-Net, our proposed method has better performance on three datasets.

Retinal Vessel Segmentation Using Supervised Classification Based on Multi-Scale Vessel Filtering and Gabor Wavelet

2015 ◽  
Vol 5 (7) ◽  
pp. 1571-1574 ◽  
Author(s):  
Songyuan Tang ◽  
Tong Lin ◽  
Jian Yang ◽  
Jingfan Fan ◽  
Danni Ai ◽  
...  
Keyword(s):  
Supervised Classification ◽  
Retinal Vessel ◽  
Vessel Segmentation ◽  
Gabor Wavelet ◽  
Multi Scale ◽  
Retinal Vessel Segmentation
Sign in / Sign up

Export Citation Format

Share Document