Table 5 The performance of the improved network vs. standard YOLOv5

From: Improving multi-scale detection layers in the deep learning network for wheat spike detection based on interpretive analysis

Method	Input size	FPS	Layers	Parameters	AP (%)
Standard YOLOv5	150 × 150	166	213	7 M	72.2
	300 × 300	166	213	7 M	89.2
	450 × 450	166	213	7 M	92.0
	600 × 600	166	213	7 M	93.0
Improved network	150 × 150	130	236	6 M	79.6
	300 × 300	130	236	6 M	90.0
	450 × 450	130	236	6 M	92.5
	600 × 600	130	236	6 M	93.5

Back to article page

ISSN: 1746-4811

Contact us

Submission enquiries: journalsubmissions@springernature.com