Automated Detection of Infection in Diabetic Foot Ulcer Images Using Convolutional Neural Network

Table 2

DFINET layer parameter details.

Layer type	Layer parameters; Kernel size (KS), number of filters (NF)
Conv_1	KS = 7 × 7, NF = 64
Max_pool_1	KS = 3 × 3, Stride = 2
Conv_2	KS = 3 × 3, NF = 64
Conv_3	KS = 3 × 3, NF = 128
Max_pool_2	KS = 3 × 3, Stride = 2
Conv_4a	KS = 3 × 3, NF = 128
Conv_4b	KS = 1 × 1, NF = 128
Max_pool_3	KS = 3 × 3, Stride = 2
Conv_5a	KS = 3 × 3, NF = 128
Conv_5b	KS = 1 × 1, NF = 128
Conv_6a	KS = 3 × 3, NF = 256
Conv_6b	KS = 1 × 1, NF = 256
Max_pool_4	KS = 3 × 3, Stride = 2
Conv_7a	KS = 3 × 3, NF = 256
Conv_7b	KS = 1 × 1, NF = 256
Max_pool_5	KS = 7 × 7, Stride = 2
FC_1	100
Dropout	Probability = 0.3
FC_1	2
Total number of parameters	14, 895, 440