The Image Differencing Change Detection task uses the common image differencing remote sensing workflow, which begins by creating a remote sensing index (such as NDVI) to use as input into the Image Difference raster function. Once the created index is differenced, the resulting image values are then thresholded to identify areas of change. Complete the following steps using the Image Differencing Change Detection task:
- Add imagery to the Map - Add the demo imagery to the map.
- Convert to Apparent Reflectance - Use the Apparent Reflectance raster function to convert raw imagery into top-of-atmosphere reflectance.
- Extract Layer for Change Detection - Use the Band Arithmetic raster function to extract a layer for change detection.
- Create Difference Image - Use the Minus raster function to create a difference image.
- Explore Symbology to Identify Change - Examine the differenced image by switching the color ramp.
- Threshold very high and very low values - Use the Remap raster function to remap the image values to 'change' and 'no change'.
- Add the Attribute Table Raster Function - Use the Attribute Table raster function to create a Change/NoChange map.
- Adjust threshold to Include More Change or Exclude False Positive Change - Adjust the thresholds as necessary to include or exclude change.