Skip to main content
AI-assisted article

This article was prepared with AI assistance based on an analysis of the current Lumina Studio source code, existing Wiki content, and relevant workflows. It has been checked against the current version, but omissions or details that become outdated may remain as the software evolves. If you find an error or have a clearer explanation, example, or suggestion, please leave a comment on this page or propose an edit on GitHub.

Image Preprocessing

Preprocessing happens before color segmentation. It first controls the internal working resolution, then applies Mean Shift smoothing to neighboring pixels.

Image Vectorizer preprocessing controls

Spatial Smoothing

The range is 0–50, with a default of 15. This is the spatial window radius used by Mean Shift:

  • Larger values suppress noise across a wider neighborhood.
  • If set too high, nearby thin lines and small structures can be smoothed together.

Color Smoothing

The range is 0–80, with a default of 25. It controls how different colors may be while still participating in the same Mean Shift aggregation:

  • Larger values consolidate similar colors and reduce fragments.
  • If set too high, nearby colors that should remain distinct can merge.

Spatial radius answers “how far should the filter look?” while color radius answers “how similar must the colors be?” The two controls work together and are not interchangeable.

Max Working Pixels

The UI range is 100,000–10,000,000, with a default of 3,000,000. If an input exceeds this total pixel count, the engine downsizes it before segmentation:

  • Lower limits reduce memory use and processing time.
  • Higher limits preserve more small detail at the cost of more computation.

Images below the limit are not enlarged by this setting. The result summary still reports the source dimensions; this value controls the internal processing resolution.

Related pages:

Submit feedback

Your feedback is sent privately to the Wiki maintainers and is not displayed publicly on this page.