Deep learning remote sensing image rural highway sandy road section extraction method
The invention discloses a deep learning remote sensing image rural highway sandy road section extraction method, which comprises the following steps of S100, selecting a remote sensing image which isacquired by a satellite and of which the resolution is within 1m, and taking a high-resolution image...
Saved in:
Main Authors | , , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
05.06.2020
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a deep learning remote sensing image rural highway sandy road section extraction method, which comprises the following steps of S100, selecting a remote sensing image which isacquired by a satellite and of which the resolution is within 1m, and taking a high-resolution image obtained after preprocessing as a data source; s200, constructing a network structure and a loss function extracted from a 'smooth return and unsmooth' road section; s300, training the 'smooth return and unsmooth' road segment extraction network by utilizing the training set, and repeatedly adjusting network parameters of the 'smooth return and unsmooth' road segment extraction network; s400, inputting the test set into the trained smooth-return unsmooth network, extracting a smooth-return unsmooth road section to verify and call back the network to achieve a better experimental result, and storing the network; and S500, inputting a remote sensing image acquired by a satellite into a network input end to identify a |
---|---|
Bibliography: | Application Number: CN202010020686 |