File size: 793 Bytes
0e3bfd3
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
e57acf8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
---
license: mit
language:
- en
metrics:
- f1
- accuracy
tags:
- unet
- fusion
- SAR
- multispectral
- LULC
- environment
pipeline_tag: image-classification
---

This is the model presented as the best-performing model to perform a land use land cover (LULC) classification based on multispectral and SAR data in the following article: Solórzano et al. 2021. Land Use Land Cover Classification with U-Net: Advantages of Combining Sentinel-1 and Sentinel-2 Imagery. Remote Sensing, 13, 3600. [https://doi.org/10.3390/rs13183600](https://doi.org/10.3390/rs13183600). The additional Github repository containing the code to obtain this model can be found in: [https://github.com/JonathanVSV/U-netR](https://github.com/JonathanVSV/U-netR).

The U-net architecture:

![unet2d.jpg](imgs/unet2d.jpg)