An Improved YOLOX Algorithm for Forest Insect Pest Detection

Table 1

Average precision performance of state-of-the-art object detection methods under different IoU thresholds on IP102.

FPN	28.10	54.93	23.30	—	—	—
SSD300	21.49	47.21	16.57	—	—	—
RefineDet	22.84	49.01	16.82	—	—	—
YOLOv3	25.67	50.64	21.79	—	—	—
Faster R-CNN	28.4	48.0	30.2	17.8	29.0	29.4
PAA	25.2	42.7	26.1	18.6	27.1	26.1
Dynamic R-CNN	29.4	50.7	30.3	14.6	25.9	30.4
TOOD	26.5	43.9	28.7	19.0	28.3	27.4
Spare R-CNN	21.1	33.2	23.8	10.2	24.3	22.0
YOLOX	31.1	52.1	32.3	23.2	32.4	32.0
Improved YOLOX	32.4	53.6	33.4	24.8	33.5	32.9