Table 1 Test accuracy on raw images with... | Emerald Publishing

Table 1

Test accuracy on raw images with 10% training data.

Accuracy	ap2or	hor2zeb	win2sum	city sc.	facades	map2sat	Ukiyoe	Van Gogh	Cezanne	Monet	average
DenseNet	79.1	95.8	67.7	93.8	99.0	78.3	99.5	97.7	99.9	89.8	89.2
XceptionNet	95.9	99.2	76.7	100.0	98.6	76.8	100.0	99.9	100.0	95.1	94.5
InceptionNet	85.0	94.8	58.8	99.4	94.0	70.5	99.8	98.8	99.9	89.9	89.1
Cozzolino2017	99.9	99.9	61.2	99.9	97.3	99.6	100.0	99.9	100.0	99.2	95.1
Auto-Spec	98.3	98.4	93.3	100.0	100.0	78.6	99.9	97.5	99.2	99.7	97.2
Nataraj2019	99.7	99.8	99.8	80.6	92.0	97.5	99.6	100.0	99.6	99.2	96.8
RGGID (6 channels)	99.2	99.8	100.0	94.4	100.0	94.1	100.0	100.0	100.0	99.4	98.7
RGGID (9 channels)	99.2	99.7	100.0	94.4	100.0	95.8	100.0	100.0	100.0	99.2	98.8
RGGID (12 channels)	99.2	99.9	100.0	95.9	100.0	95.8	100.0	100.0	100.0	99.1	99.0

Accuracy	ap2or	hor2zeb	win2sum	city sc.	facades	map2sat	Ukiyoe	Van Gogh	Cezanne	Monet	average
DenseNet	79.1	95.8	67.7	93.8	99.0	78.3	99.5	97.7	99.9	89.8	89.2
XceptionNet	95.9	99.2	76.7	100.0	98.6	76.8	100.0	99.9	100.0	95.1	94.5
InceptionNet	85.0	94.8	58.8	99.4	94.0	70.5	99.8	98.8	99.9	89.9	89.1
Cozzolino2017	99.9	99.9	61.2	99.9	97.3	99.6	100.0	99.9	100.0	99.2	95.1
Auto-Spec	98.3	98.4	93.3	100.0	100.0	78.6	99.9	97.5	99.2	99.7	97.2
Nataraj2019	99.7	99.8	99.8	80.6	92.0	97.5	99.6	100.0	99.6	99.2	96.8
RGGID (6 channels)	99.2	99.8	100.0	94.4	100.0	94.1	100.0	100.0	100.0	99.4	98.7
RGGID (9 channels)	99.2	99.7	100.0	94.4	100.0	95.8	100.0	100.0	100.0	99.2	98.8
RGGID (12 channels)	99.2	99.9	100.0	95.9	100.0	95.8	100.0	100.0	100.0	99.1	99.0