Table 4 State of art analysis
From: Retinal vessel segmentation using multi scale feature attention with MobileNetV2 encoder
Reference number | Technique used | Dataset used | Accuracy | Jaccard/IoU |
|---|---|---|---|---|
GLCAA | DRIVE | 0.9603 | 0.5928 | |
MPCCN | DRIVE | 0.9738 | 0.8185 | |
ANSAN-Infused Retinal Vessel Segmentation | DRIVE | 0.88 | 0.7764 | |
Morphology Cascaded Features and Supervised Learning | DRIVE | 0.9747 | 0.6199 | |
Spatial Attention U-Net (SA-UNet) | DRIVE | 0.9583 | 0.7011 | |
Genetic U-Net | DRIVE | 0.9704 | 0.6783 | |
Spider U-Net (LSTM for 3D Segmentation) | DRIVE | 0.9697 | 0.6812 | |
MRU-Net (U-Net Variant) | DRIVE | 0.9837 | 0.7291 | |
M2U-Net | DRIVE | 0.963 | - | |
Context Encoder Network (CE-Net) | DRIVE | 0.9523 | 0.81 | |
LUVS-Net (Lightweight U-Net) | DRIVE | 0.9578 | 0.7955 | |
Proposed model | MSFAUMobileNet Model | DRIVE | 0.9999 | 0.9994 |
