Skip to main content

Table 1 The summary Of ResNet50 layered architecture. Architecture

From: Review of deep learning methods for remote sensing satellite images classification: experimental survey and comparative analysis

Block

Units (n)

Operators (F)

Resolutions (\({H} \times {W}\))

Channels

Block A

1

Conv, ReLu, MaxPool, BatchNorm

\(112\times 112\)

64

Block B

15

Conv, ReLu, BatchNorm

\(56\times 56, 28 \times 28,14 \times 14, 7\times 7\)

\(256,\ldots ,2048\)

Block C

10

Conv, BatchNorm

\(56\times 56, 28 \times 28,14 \times 14, 7 \times 7\)

64, 128, ..., 2048

Block D

12

Conv, Relu, BatchNorm, AdaptiveMaxPool, AdaptAvgPool

\(7\times 7\)

2048

Block E

1

Linear, Relu, BatchNorm

\(7\times 7\)

512

Block F

1

Linear

\(7\times 7\)

10