Analyze raster data and imagery
A practical guide to raster values, imagery indices, terrain, classification, and zonal summaries.
A raster stores values in cells. Depending on its source, the values may represent elevation, reflectance, temperature, a category, or another measurement. Inspect the layer's metadata and legend before interpreting those values.
Imagery and indices
You can ask GriidAi to compare bands, create a color composite, or calculate an index such as NDVI when the selected product contains the needed data. Band numbers, scaling, units, and quality masks depend on the product. Ask the assistant to inspect the image first if you are not sure which bands to use.
Terrain and raster calculations
A digital elevation model can support terrain products such as slope, aspect, or shaded relief. For a custom raster calculation, state the inputs, units, output, and any threshold or classification you want. Ask how no-data cells and resampling are handled when those choices matter.
Summaries by area
A zonal summary combines raster cells with vector zones. Specify the statistic (for example, mean or maximum), the raster band, and the polygons. Cell size, alignment, no-data cells, and boundary handling can affect the answer. Review the output fields and units.
Categories and display
A classified raster stores category codes or values. Use its legend to understand the categories; colors are only a way to display them. Check no-data values separately from valid classes before counting or measuring pixels.
Need assistance with this workflow?
Our geospatial engineering team is available to help with custom datasets, STAC queries, and algorithms.