Research Article

An Expert System for COVID-19 Infection Tracking in Lungs Using Image Processing and Deep Learning Techniques

Table 4

VGGNet: deep neural network layers.

Layer (type)Output shapeParameter #

conv2d (Conv2D)(None, 444, 444, 32)2432
max_pooling2d (MaxPooling2D)(None, 222, 222, 32)0
conv2d_1 (Conv2D)(None, 220, 220, 64)18496
max_pooling2d_1 (MaxPooling2)(None, 110, 110, 64)0
dropout (dropout)(None, 110, 110, 64)0
conv2d_2 (Conv2D)(None, 108, 108, 128)73856
max_pooling2d_2 (MaxPooling2)(None, 54, 54, 128)0
dropout_1 (Dropout)(None, 54, 54, 128)0
conv2d_3 (Conv2D)(None, 52, 52, 512)590336
max_pooling2d_3 (MaxPooling2)(None, 26, 26, 512)0
dropout_2 (Dropout)(None, 26, 26, 512)0
conv2d_4 (Conv2D)(None, 24, 24, 512)2359808
conv2d_5 (Conv2D)(None, 22, 22, 128)589952
conv2d_6 (Conv2D)(None, 20, 20, 64)73792
max_pooling2d_4 (MaxPooling2)(None, 10, 10, 64)0
dropout_3 (dropout)(None, 10, 10, 64)0
flatten (flatten)(None, 6400)0
dense (dense)(None, 4096)26218496
dense_1 (dense)(None, 1024)4195328
dropout_4 (dropout)(None, 1024)0
dense_2 (dense)(None, 3)3075
Total parameters: 34125571; trainable parameters: 34125571; nontrainable parameters: 0