Road surface crack detection method based on edge reconstruction network
The invention discloses a road surface crack detection method based on an edge reconstruction network, and the method comprises the steps: constructing a binary cross entropy loss function, an edge loss function and a detail loss function, and taking the sum of the three functions as a total loss fu...
Saved in:
Main Authors | , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
28.05.2024
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a road surface crack detection method based on an edge reconstruction network, and the method comprises the steps: constructing a binary cross entropy loss function, an edge loss function and a detail loss function, and taking the sum of the three functions as a total loss function; using an edge loss function to strengthen extraction of boundary features by edge branches, using a detail loss function to strengthen extraction of boundary features by space branches, and using a binary cross entropy loss function to strengthen pixel classification; in the training stage, training set images are input into the network, loss calculation is carried out on a prediction result of the network through a loss function, and network parameters are optimized through back propagation and iterative training; and after the loss is stable, inputting the test set image into the trained neural network to obtain a final detection result. The road crack detection generalization can be effectively enhanced, |
---|---|
Bibliography: | Application Number: CN202410172558 |