Skip to main content
. 2021 Jan 14;11:1447. doi: 10.1038/s41598-021-81216-5

Table 3.

Model performance evaluation with 0.25 dataset under 416 × 416 pixels’ resolution.

Methods Activation FDL Precision (%) Recall (%) F1 (%) AP (%)
YOLOv3 Leaky  × 3 96.3 95.9 96.1 97.1
YOLOv4 Mish + SPP  × 3 98.0 100.0 99.0 99.0
YOLO-Tomato-A Leaky  × 3 96.1 98.5 97.3 98.3
YOLO-Tomato-B Mish  × 1 95.6 98.9 97.2 99.2
YOLO-Tomato-C Mish + SPP  × 2 96.4 99.5 97.9 99.4