iResSENet: An Accurate Convolutional Neural Network for Retinal Blood Vessel Segmentation
In this paper, we propose iResSENet, a novel deep learning-based image segmentation model based on U-Net architecture. The proposed method enhances U-Net in three aspects. It replaces the encoder blocks with residual connections in addition to 1×1 convolutional layers and channel-based attention. The proposed method was applied for segmentation tasks in retinal blood vessels. The experimental results show that the proposed method is significantly superior compared to existing methods on several standard benchmark datasets.
PDF AbstractResults from the Paper
Submit
results from this paper
to get state-of-the-art GitHub badges and help the
community compare results to other papers.