MixerNet-SAGA A Novel Deep Learning Architecture for Superior Road Extraction in High-Resolution Remote Sensing Imagery

MixerNet-SAGA A Novel Deep Learning Architecture for Superior Road Extraction in High-Resolution Remote Sensing Imagery

In this study, we address the limitations of current deep learning models in road extraction tasks from remote sensing imagery. We introduce MixerNet-SAGA, a novel deep learning model that incorporates the strengths of U-Net, integrates a ConvMixer block for enhanced feature extraction, and includes...

Full description

Bibliographic Details
Main Authors: Wei Wu, Chao Ren, Anchao Yin, Xudong Zhang
Format: Article
Language:English
Published: MDPI AG 2023-09-01
Series:Applied Sciences
Subjects:
high-resolution remote sensing imagery
road extraction
MixerNet-SAGA
ConvMixer blocks
scaled attention mechanisms
deep learning architectures
Online Access:https://www.mdpi.com/2076-3417/13/18/10067

Internet

https://www.mdpi.com/2076-3417/13/18/10067

Similar Items