TABLE 7.

Comparison of the results from ImageNet and PlantNet pre-trained backbones in the cascade detector.

Class	Average precision of ImageNet pre-trained backbone	Average precision of PlantNet
Angular leafspot	88.41%	94.55%
Anthracnose fruit rot	90.23%	96.01%
Gray mold	90.17%	94.52%
Leaf blight	89.81%	91.82%
Leaf spot	85.16%	87.83%
Powdery mildew fruit	84.01%	86.36%
Powdery mildew leaf	88.59%	90.48%
mAP	88.05%	91.65%