An improved Image Difference Metric


Gjøvik University College,
London College of Communication
University of the Arts London

Project supervisors:

Prof. Tony Johnson (Director of studies),
Dr. Phil Green (co-supervisor),
Dr. Jon Y Hardeberg (local supervisor),
Prof. Lindsay MacDonald (Advisor),
PhD student: Eriko Bando

Project outline

There is as yet no clear recommendation for quantifying visual image difference between the original and reproductions of photo realistic images because of the complexity of these multiple factors. My research aims 1. To develop an image difference metric which takes into account attributes of colour perception including lightness, chroma and hue, as well as other factors such as image content, substrate and media. 2. Implement and evaluate the metric in a graphic arts proofing context. 3. Indicate how the metric can be extended to incorporate other quality parameters.

The research is funded by : The Research Council of Norway and Norwegian companies through the Quality Assured Content Management project, and Gjøvik University College.

News archive