Table 2 Different ablation experimental results of GEW-YOLO model on Seaships dataset. y/n represents none.
From: Multi-ship detection and classification with feature enhancement and lightweight fusion
Model | GSConvns | ESSE | Wise-IoU | Precision(%) | Recall(%) | mAP@0.5(%) |
|---|---|---|---|---|---|---|
YOLOv8 | n | n | n | 94.5 | 95.5 | 97.1 |
YOLOv8+GSConvns | y | n | n | 95.6 | 93.8 | 97.6 |
YOLOv8+ESSE | n | y | n | 97.4 | 96.4 | 98.3 |
YOLOv8+Wise-IoU | n | n | y | 95.9 | 94 | 98.4 |
y/n GSConvns | n | y | y | 97.4 | 97 | 98.9 |
y/n ESSE | y | n | y | 95.1 | 95 | 98.1 |
y/n Wise-IoU | y | y | n | 96.7 | 96.1 | 98.7 |
GEW-YOLO(ours) | y | y | y | 97.6 | 98.1 | 99.1 |
