Table 1.
Layer | Type | Input | Kernel | Output |
---|---|---|---|---|
1 | Convolution layer | 32 × 32 × 1 | 6 × 6 | 28 × 28 × 36 |
2 | Max pooling | 28 × 28 × 36 | 3 × 3 | 24 × 24 × 64 |
3 | Convolution layer | 24 × 24 × 64 | 6 × 6 | 20 × 20 × 64 |
4 | Max pooling | 20 × 20 × 64 | 3 × 3 | 12 × 12 × 64 |
5 | Fully connected layer | 12 × 12 × 64 | 5 × 5 | 8 × 8 × 64 |
6 | Fully connected layer | 8 × 8 × 64 | 2 × 2 | 2 × 1 |
7 | Softmax layer | 2 × 1 result | - | Identification output |