NDVI Calculator

Block type: PROCESSING

This block takes photos from an optical satellite and calculates Normalized Difference Vegetation Index (NDVI). It produces georeferenced raster output, which can be inspected visually or used for further processing. The output can be opened in GIS tools like QGIS or ArcGIS.

Supported parameters

• colormap - the colormap, which will be used in rendering. See section Colormap for valid values.

• satellite - the name of the block providing input data. See section Supported data blocks for valid values.

Example parameters using Sentinel-2 Level 2 (BOA) AOI clipped block as data source and then rendering NDVI with qz_ndvi colormap:

{
  "sentinelhub-s2-aoiclipped:1": {
    "time": "2018-05-07T16:47:48+00:00/2018-05-09T16:47:49+00:00",
    "limit": 1,
    "intersects": {
      "type": "Polygon",
      "coordinates": [
        [
          [
            18.726196,
            53.473489
          ],
          [
            18.720531,
            53.479312
          ],
          [
            18.708215,
            53.476043
          ],
          [
            18.719931,
            53.468431
          ],
          [
            18.73332,
            53.466183
          ],
          [
            18.739586,
            53.471854
          ],
          [
            18.726196,
            53.473489
          ]
        ]
      ]
    },
    "zoom_level": 14
  },
  "qzsolutions.ndvi:1": {
    "colormap": "qz_ndvi",
    "satellite": "Sentinel-2 Level 2 (BOA) AOI clipped"
  }
}

Supported data blocks

Currently this block supports input from data blocks:

Landsat-8 Level 1 (TOA) AOI clipped

To see this block on up42 please click here.

Sample block parameters when using Landsat-8 Level 1 (TOA) AOI clipped data block:

{
  "colormap": "qz_ndvi",
  "satellite": "Landsat-8 Level 1 (TOA) AOI clipped"
}

Pléiades Download

To see this block on up42 please click here.

Sample block parameters when using Pléiades Download data block:

{
  "colormap": "qz_ndvi",
  "satellite": "Pleiades Download"
}

Warning

Please notice that satellite parameter does not contain ‘é’ character but ‘e’.

Sentinel-2 L1C MSI Full Scenes

To see this block on up42 please click here.

Sample block parameters when using Sentinel-2 L1C MSI Full Scenes data block:

{
  "colormap": "qz_ndvi",
  "satellite": "Sentinel-2 L1C MSI Full Scenes"
}

Sentinel-2 Level 2 (BOA) AOI clipped

To see this block on up42 please click here.

Sample block parameters when using Sentinel-2 Level 2 (BOA) AOI clipped data block:

{
  "colormap": "qz_ndvi",
  "satellite": "Sentinel-2 Level 2 (BOA) AOI clipped"
}

SPOT 6/7 Download

To see this block on up42 please click here.

Sample block parameters when using SPOT 6/7 Download data block:

{
  "colormap": "qz_ndvi",
  "satellite": "SPOT 6/7 Download"
}

Output format

The block produces a feature collection with up42.data_path property containing a name of the result file.

Contents of data.json file returned by a block configured in the previous example:

{
  "type": "FeatureCollection",
  "features": [
    {
      "type": "Feature",
      "bbox": [
        18.69873046875,
        53.46189043285912,
        18.74267578125,
        53.48804553605621
      ],
      "geometry": {
        "type": "Polygon",
        "coordinates": [
          [
            [
              18.720703125,
              53.46189043285912
            ],
            [
              18.69873046875,
              53.46189043285912
            ],
            [
              18.69873046875,
              53.47496999954854
            ],
            [
              18.69873046875,
              53.48804553605621
            ],
            [
              18.720703125,
              53.48804553605621
            ],
            [
              18.74267578125,
              53.48804553605621
            ],
            [
              18.74267578125,
              53.47496999954854
            ],
            [
              18.74267578125,
              53.46189043285912
            ],
            [
              18.720703125,
              53.46189043285912
            ]
          ]
        ]
      },
      "properties": {
        "up42.data_path": "6e25951d-25f0-4cec-ac27-2687f360ccf9.tiff",
        "acquisitionDate": "2018-05-08T09:50:28"
      }
    }
  ]
}

Contents of file 6e25951d-25f0-4cec-ac27-2687f360ccf9.tiff

Legend is in section RGB Output.

Capabilities

Input

This block has such input_capabilities:

"input_capabilities": {
  "raster": {
    "up42_standard": {
      "format": {
        "or": [
          "SAFE",
          "GTiff",
          "DIMAP"
        ]
      },
      "sensor": {"or": ["Landsat8", "Sentinel2", "SPOT", "Pleiades"]},
      "bands": ["red", "nir"]
    }
  }
}

Warning

Please ensure that input block is listed under Supported data blocks section.

Output

This block has such output_capabilities:

"output_capabilities": {
  "raster": {
    "up42_standard": {
      "format": "GTiff",
      "dtype": {"or": ["uint8", "float"]},
      "resolution": ">"
    },
    "custom": {
      "bands": [
        "ndvi"
      ]
    }
  }
}

If colormap is set to id then output type is float. Otherwise it’s uint8.

Colormap

NDVI is always in \([-1, 1]\) range. There are two possible output types: RGB output and raw output. The output type is controlled with colormap parameter (see Colormap).

RGB Output

We can visualize NDVI as RGB GeoTIFF file. Colormap assigns color to each possible index value.

A legend for commonly used color maps is in the figure below.

Commonly used colormaps

This block supports any predefined Matplotlib colormaps.

Additionally, you can use a qz_ndvi colormap, designed specifically for NDVI visualization. It shows water as shades of blue, rocks and soils show up as shades of brown and vegetation as shades of green.

Set colormap parameter to requested colormap name (see Colormap).

Raw Output

It is possible to get raw output, i.e. single band GeoTIFF file with float32 matrix containing NDVI values. Set colormap parameter to id (see Colormap).

{
  "colormap": "id",
  "satellite": "SPOT 6/7 Download"
}

Such files can be manually colorized in QGIS with singleband pseudocolor. It is possible to read NDVI value for particular pixel with Identify Features tool.

Other usage of such files is further processing in Python or as an input for other up42 processing blocks.