Application of Deep Learning in Integrated Pest Management: A Real-Time System for Detection and Diagnosis of Oilseed Rape Pests

Table 5

Validation set results for the model with and without data augmentation.

Method	Athalia rosae japanensis	Creatonotus transiens	Entomoscelis adonidis	Entomoscelis suturalis
Without data augmentation	0.8218	0.586	0.6176	0.7555
Data augmentation	0.7858	0.6106	0.8593	0.9167
Method	Hellula undalis	Lipaphis erysimi	Mamestra brassicae	Meligethes aeneus
Without data augmentation	0.7738	0.3772	0.967	0.2247
Data augmentation	0.842	0.5914	0.9577	0.301
Method	Phyllotreta striolata	Pieris rapae	Plutella xylostella	Psylliodes punctifrons	mAP
Without data augmentation	0.6275	0.9111	0.8767	0.554	0.6744
Data augmentation	0.7759	0.7728	0.8062	0.6808	0.7417