Figure 3
From: Detecting common coccinellids found in sorghum using deep learning models
Data flow diagram. Coccinellid images are collected from iNaturalist and labeled using Labelbox. The set of images is split between train, dev and test subsets. YOLO and Faster R-CNN-FPN models trained and fine-tuned on the train and dev subsets, respectively. The performance of each model is estimated on the test subset in terms of average precision (AP) metrics.
