This event has ended. View the official site or create your own event → Check it out
This event has ended. Create your own
View analytic
Friday, May 31 • 8:30am - 9:00am
(Paintings + Research and Technical Studies) Color and Spectral Archiving Using a Dual-RGB Imaging System

Sign up or log in to save this to your schedule and see who's attending!

Digital technologies present conservators with numerous new tools on many fronts including core activities like documentation. Often, though, the cost and complexity of the new technology is a barrier to its full implementation by the field. This paper will present the development and refinement of a multispectral imaging system (Dual-RGB) that uses off-the-shelf hardware and simple custom software, thereby reducing cost and complexity, to demonstrably improve the color accuracy for conservation documentation through a spectral estimation technique.

This system consists of a standard digital camera of reasonably high spatial resolution, two custom filters and standard studio lights for capture of the images. The camera has the IR filter removed to extend sensitivity. Processing of the images is done with a GUI based program generated in Matlab and executable on any OS.

Fundamentally the system increases the number of channels captured from the standard 3 rgb channels to 6 by placing the two filters in the optical path and capturing images with each filter, hence the name Dual-RGB. First a white board is imaged with the filters so that the object images can be flat fielded to remove lighting artifacts and further improve the spectral estimation. A target such as a ColorChecker Classic with known spectral measurements is then imaged by the system to calculate a transformation matrix. This matrix will then be used to derive the spectral estimation of the object being imaged, which is also imaged with the two filters. Finally the object is captured with the two filters after which the image processing steps are done. The calibration images of white board and known target can be used for processing multiple objects provided the camera and lighting set up does not change.

Four computational steps are required for the Dual-RGB approach: flat fielding, image registration, colorimetric processing, and spectral processing. Flat fielding includes spatial low-pass filtering, effectively removing noise artifacts from the sensor and smoothing any texture or dirt of the diffuse white board. The image registration is calculated at the sub-pixel level. The archived image is a nine-channel ProPhotoRGB Tiff file. The first three channels are ProPhotoRGB encoded data. Channels 4-6 are image data from filter 1 and channels 7-9 are image data from filter 2. Channel 4-9 data have been flat fielded, registered, rescaled and encoded nonlinearly using the sRGB standard, and quantized to 16 bits.

The colorimetric and spectral processing achieves high color accuracy, reasonable spectral accuracy, and minimal noise propagation . The average DeltaE2000 for a color chart is less than 1 versus a range of 1.3 to 5.8 using the standard capture for a range of medium format digital camera backs. Perhaps more importantly the maximum error of the Dual-RGB system is 8.3 versus a range of 15 to 32 with the standard capture.

i R. S. Berns and S. Smith, “Analysis of color management default camera profiles for museum imaging applications,” IS&T Archiving 2012, 111-115 (2012).


Roy S. Berns

Richard S. Hunter Professor in Color Science, Appearance, and Technology and Director of the Munsell Color Science Laboratory, Center for Imaging Science, Rochester Institute of Technology
Prior to earning his Ph.D. from the University of Arizona in 2006, R.G. Erdmann started a science and engineering software company and worked extensively on multiscale materials modeling at Sandia National Laboratories in both Albuquerque, NM and Livermore, CA. He subsequently joined the faculty at the University of Arizona as a joint hire between the Department of Materials Science and Engineering and the Program in Applied Mathematics. His... Read More →

Jim Coddington

Chief Conservator, Museum of Modern Art
Jim Coddington is the Agnes Gund Chief Conservator at the Museum of Modern Art, New York. He has a B.A. from Reed College and an M.S. from the University of Delaware/Winterthur Museum Conservation Program.

Friday May 31, 2013 8:30am - 9:00am
JW Marriott White River Ballroom F 10 S West St Indianapolis, IN 46204

Attendees (30)