Skip to main content

Table 3 Accuracy and mean average precision (mAP) scores of detecting four species of white flowers includingAndrosace umbellata (Lour.) Merr., Bidens pilosa L., and Trifolium repens L. and Fragaria × ananassa based on deep learning methods of convolutional neural network on the training and test dataset, respectively

From: Deep convolutional neural network for automatic discrimination between Fragaria × Ananassa flowers and other similar white wild flowers in fields

Method

Training set

Test set

Accuracy (%)

mAP

Accuracy (%)

mAP

SIFT + SVM

82.9

0.960

55.6

0.794

PHOG + SVM

78.3

0.900

63.1

0.744

CNNs

99.2

0.983

95.0

0.974