Skip to content

darts_preprocessing.calculate_ndvi

Calculate NDVI from an xarray Dataset containing spectral bands.

Example 
ndvi_data = calculate_ndvi(planet_scene_dataset)

Parameters:

  • planet_scene_dataset (xarray.Dataset) –

    The xarray Dataset containing the spectral bands, where the bands are indexed along a dimension (e.g., 'band'). The Dataset should have dimensions including 'band', 'y', and 'x'.

  • nir_band (str, default: 'nir' ) –

    The name of the NIR band in the Dataset (default is "nir"). This name should correspond to the variable name for the NIR band in the 'band' dimension. Defaults to "nir".

  • red_band (str, default: 'red' ) –

    The name of the Red band in the Dataset (default is "red"). This name should correspond to the variable name for the Red band in the 'band' dimension. Defaults to "red".

Returns:

  • xarray.Dataset –

    xr.Dataset: A new Dataset containing the calculated NDVI values. The resulting Dataset will have dimensions (band: 1, y: ..., x: ...) and will be named "ndvi".

Notes

NDVI (Normalized Difference Vegetation Index) is calculated using the formula: NDVI = (NIR - Red) / (NIR + Red)

This index is commonly used in remote sensing to assess vegetation health and density.

Source code in darts-preprocessing/src/darts_preprocessing/engineering/indices.py 
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
@stopuhr.funkuhr("Calculating NDVI", printer=logger.debug, print_kwargs=["nir_band", "red_band"])
def calculate_ndvi(planet_scene_dataset: xr.Dataset, nir_band: str = "nir", red_band: str = "red") -> xr.Dataset:
    """Calculate NDVI from an xarray Dataset containing spectral bands.

    Example:
        ```python
        ndvi_data = calculate_ndvi(planet_scene_dataset)
        ```

    Args:
        planet_scene_dataset (xr.Dataset): The xarray Dataset containing the spectral bands, where the bands are indexed
            along a dimension (e.g., 'band'). The Dataset should have dimensions including 'band', 'y', and 'x'.
        nir_band (str, optional): The name of the NIR band in the Dataset (default is "nir"). This name should
            correspond to the variable name for the NIR band in the 'band' dimension. Defaults to "nir".
        red_band (str, optional): The name of the Red band in the Dataset (default is "red"). This name should
            correspond to the variable name for the Red band in the 'band' dimension. Defaults to "red".

    Returns:
        xr.Dataset: A new Dataset containing the calculated NDVI values. The resulting Dataset will have
            dimensions (band: 1, y: ..., x: ...) and will be named "ndvi".


    Notes:
        NDVI (Normalized Difference Vegetation Index) is calculated using the formula:
            NDVI = (NIR - Red) / (NIR + Red)

        This index is commonly used in remote sensing to assess vegetation health and density.

    """
    # Calculate NDVI using the formula
    nir = planet_scene_dataset[nir_band].astype("float32")
    r = planet_scene_dataset[red_band].astype("float32")
    ndvi = (nir - r) / (nir + r)

    # Scale to 0 - 20000 (for later conversion to uint16)
    ndvi = (ndvi.clip(-1, 1) + 1) * 1e4
    # Make nan to 0
    ndvi = ndvi.fillna(0).rio.write_nodata(0)
    # Convert to uint16
    ndvi = ndvi.astype("uint16")

    ndvi = ndvi.assign_attrs({"data_source": "planet", "long_name": "NDVI"}).to_dataset(name="ndvi")
    return ndvi