基于轻量高频Transformer与特征互补融合的视网膜血管分割

梁礼明,王成斌,钟奕,陈林俊,吴健

Retinal vessel segmentation based on lightweight high-frequency Transformer and feature complementary fusion

Liming LIANG,Chengbin WANG,Yi ZHONG,Linjun CHEN,Jian WU

表 1 不同算法在3个数据集上的血管分割结果对比

Tab.1 Comparison of vessel segmentation results by different algorithms on three datasets

数据集 模型 ACC/% SE/% SP/% AUC/% F1/% DRIVE U-Net 97.08 78.57 98.86 98.81 82.53 Attention U-Net 97.10 79.42 98.80 98.82 82.78 FR-UNet 97.06 80.10 98.69 98.78 82.69 GT-DLA-dsHFF 97.10 80.15 98.73 98.80 82.91 LFF-Net 97.12 80.23 98.74 98.83 82.99 STARE U-Net 97.52 79.26 99.02 99.01 82.98 Attention U-Net 97.55 78.88 99.01 99.05 83.08 FR-UNet 97.49 80.36 98.90 99.03 83.01 GT-DLA-dsHFF 97.56 79.98 99.01 99.09 83.37 LFF-Net 97.62 80.48 99.02 99.12 83.73 CHASE_DB1 U-Net 97.46 80.90 98.57 98.86 80.09 Attention U-Net 97.60 80.65 98.74 98.97 80.96 FR-UNet 97.41 81.19 98.50 98.75 79.83 GT-DLA-dsHFF 97.61 80.72 98.74 99.00 80.97 LFF-Net 97.65 81.30 98.75 98.99 81.37