The growth of automatic layout capabilities for publications such as photo books and image sharing websites enables consumers to create personalized presentations without much experience or the use of professional page design software. Automated color correction of images has been well studied over the years, but the methodology for determining how to correct images has almost exclusively considered images as independent indivisible objects. In modern documents, such as photo books or web sharing sites, images are automatically placed on pages in juxtaposition to others and some images are automatically cropped. Understanding how color correction preferences are impacted by complex arrangements has become important. A small number of photographs taken under a variety illumination conditions were presented to observers both individually and in combinations. Cropped and uncropped versions of the shots were included. Users had opportunities to set preferred color balance and chroma for the images within the experiment. Analyses point toward trends indicating a preference for higher chroma for most cropped images in comparison to settings for the full spatial extent images. It is also shown that observers make different color balance choices when correcting an image in isolation versus when correcting the same image in the presence of a second shot taken under a different illuminant. Across 84 responses, approximately 60% showed the tendency to choose image white points that were further from the display white point when multiple images from different taking illuminants were simultaneously presented versus when the images were adjusted in isolation on the same display. Observers were also shown to preserve the relative white point bias of the original taking illuminants.
Library of Congress Subject Headings
Color photography--Digital techniques; Image processing--Digital techniques; Photography--Retouching
Department, Program, or Center
Chester F. Carlson Center for Imaging Science (COS)
Fredericks, Erin P., "Preferred color correction for mixed taking-illuminant placement and cropping" (2009). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus