Table 1 Parameterizations of the improved Faster R-CNN based ECA mechanism

From: Automatic defect detection in infrared thermal images of ancient polyptychs based on numerical simulation and a new efficient channel attention mechanism aided Faster R-CNN model

Parameter name

Total training steps

Basic learning rate

Value

37000

0.0001

Parameter name

Parameters

\(\epsilon\)

Impulse setting

Value

\({10}^{-8}\)

0.9

Back to article page