Illumination problem
Data-Driven Parameter Choice for Illumination Artifact Correction of Digital Images
Data-Driven Parameter Choice for Illumination Artifact Correction of Digital Images
Illumination artifacts arise in a variety of digital images such as digital photography, digital microscopy or Magnetic Resonance images. Illumination artifacts consist of “smooth” variations in color/gray level intensity which are visible in the acquired image but are not present in the original scene. We propose a new procedure for image illumination correction with data-driven parameter choice.
Results of a sample of 50 images are available here.