Predicted Anchor Region Proposal with Balanced Feature Pyramid for License Plate Detection in Traffic Scene Images

Table 1

The architectures of ResNet-34 for ImageNet.

Layer name	Kernel size	Output size
Conv1	7 × 7 × 64, stride 2	112 × 112
Conv2	3 × 3 max pool, stride 2	56 × 56
	Conv3			28 × 28
Conv4		14 × 14
Conv5		7 × 7

Downsampling is performed by conv3-1, conv4-1, and conv5-1 with a stride of 2.