Standard Deviation Map
Generates a local spatial-variation map for every spectral band.
Overview
Generates a local spatial-variation map for every spectral band.
Current IDCubePro Implementation
IDCubePro applies MATLAB stdfilt independently to each spectral band using a 3 x 3 spatial neighborhood.
Each output pixel therefore represents local spatial intensity variation around that pixel within the corresponding band.
Higher values indicate stronger local spatial variation.
Lower values indicate more locally uniform image regions.
- Inspecting local texture or spatial heterogeneity.
- Identifying boundaries or rapidly changing image regions.
- Comparing spatial variation across spectral bands.
Important
This is a spatial standard-deviation calculation within each band.
It is not the standard deviation across the spectral dimension.
Strong noise, edges, texture, and real heterogeneous structure can all produce elevated local standard deviation.
Workflow
1. Load a hyperspectral dataset.
2. Open Filtering & Enhancement.
3. Select Standard Deviation Map.
4. Allow processing to complete.
5. Inspect bright and dark regions in the resulting maps.
6. Compare with the original image before interpretation.