Table 7 AP outcome of the FFODL-VDCAI approach with other models on the PSU dataset.
From: Flying foxes optimization with reinforcement learning for vehicle detection in UAV imagery
Average precision by object size (PSU dataset) | |||
|---|---|---|---|
Car size | Small | Medium | Large |
Faster R-CNN(Inceptionv2) | 0.75 | 0.46 | 0.53 |
Faster R-CNN(Resnet50) | 0.68 | 0.52 | 0.48 |
YOLO-v3 (320 × 320) | 0.89 | 0.50 | 0.62 |
YOLO-v4 (320 × 320) | 0.95 | 0.56 | 0.72 |
FFODL-VDCAI | 0.98 | 0.74 | 0.82 |
