Table 1 Comparison of weed identification results using different models.

Model	Model Size(MB)	Accuracy	Precision	Recall	F1	Inference Time(ms)
ShuffleNet	18.9	0.9741	0.9838	0.9294	0.9558	105
MobileNetv1	17.3	0.9423	0.9642	0.8806	0.8724	119
MobileNetv2	14.9	0.9700	0.9730	0.9378	0.9551	183
MobileNetv3	11.9	0.8790	0.8168	0.9265	0.8682	174
ResNet-50	112.9	0.9538	0.9340	0.9698	0.9516	459
VGG-16	882.9	0.9294	0.9091	0.9535	0.9308	589
DenseNet-161	283	0.9775	0.9652	0.9848	0.9749	782
MobileNetV3-Large	23.2	0.9631	0.9521	0.9516	0.9518	264
EfficientNet-Lite	51.4	0.9569	0.9481	0.9421	0.9450	325
GhostNet	21.6	0.9341	0.9431	0.9221	0.9324	214
Our Model	12.2	0.9856	0.9726	0.9979	0.9851	89

Quick links

Search