Processing classified output
In the classified output, some misclassified isolated pixels or small regions of pixels may exist. This gives the output a "salt and pepper" or speckled appearance. Post-classification processing refers to the process of removing the noise and improving the quality of the classified output.
The ArcGIS Spatial Analyst extension provides a set of generalization tools for the post-classification processing task. This task involves three steps. Each step is based on a Spatial Analyst tool from the Generalization toolset.
- Filtering the classified output.
This step removes the isolated pixels or noise from the classified output. The Majority Filter tool is used.
- Smoothing class boundaries and clumping classified output.
This step smooths the ragged class boundaries and clumps the classes. The Boundary Clean tool is used.
- Generalizing classified output by removing small isolated regions.
This step reclassifies small isolated regions of pixels to the nearest classes. The Region Group, Set Null and Nibble tools are used.