DeepLumina: A Method Based on Deep Features and Luminance Information for Color Texture Classification

Table 3

CNN parameters used for feature map generation from the Pretrained Models.

CNN parameters	ResNet101	MobileNet	VGG19	InceptionResNetv2
Convlayer name	conv1	conv_1	conv1_1	conv2d_1
No.of layers	347	155	47	825
Input image size	[224,224,3]	[224,224,3]	[224,224,3]	[299,299,3]
Filter size	7 × 7	3 × 3	3 × 3	3 × 3
No of filters	64	32	64	32
Stride	[2,2]	[2,2]	[1,1]	[2,2]
Dilation factor	[1,1]	[1,1]	[1,1]	[1,1]
Padding size	[3,3,3,3]	[0,1,0,1]	[1,1,1,1]	[0,0,0,0]
Weight initialization	Glorot	Glorot	Glorot	Glorot
Feature layer	fc1000	Logits	fc8	Predictions