Land Cover Classifier for Pléiades/SPOT
Classifies imagery into discrete land cover classes.
Analytics platform discontinued after January 31, 2024
The current analytics platform will be discontinued after January 31, 2024, and will be replaced by new advanced processing functionalities. This change will affect projects, workflows, jobs, data blocks, processing blocks, and custom blocks. For more information, see the blog post.
Description
A processing block that assigns each of the input image pixels to a discrete land cover class.
See this block on the marketplace.
Compatibility
Catalog collections
Use in combination with Processing from Storage.
How it works
Parameter | Overview |
---|---|
nclasses | integer / required The number of land cover classes. The allowed values:
4 . |
Land cover classes
Class | Overview |
---|---|
1 | Water |
2 | High vegetation (including trees) |
3 | Low vegetation (including bushes and grass) |
4 | Barren land |
5 | Urban (including roads and buildings) |
Accuracy assessment
Accuracy metrics | With 4 classes | With 5 classes |
---|---|---|
Accuracy | 0.75 | 0.64 |
Jaccard | 0.22 | 0.51 |
Capabilities
Input
raster
up42_standard | |
---|---|
bands | [
"red",
"green",
"blue",
"nir"
] |
dtype | uint16 |
format | GTiff |
sensor | {
"or": [
"Pleiades",
"SPOT"
]
} |
processing_level | l2 |
Output
raster
up42_standard | |
---|---|
bands | [
"landcover"
] |
dtype | uint8 |
format | > (propagated) |
sensor | > (propagated) |
resolution | > (propagated) |
To know more please check the block capabilities specifications.