Evaluating CATs as Predictors of Observer Adjustments in Softcopy Fine Art Reproduction

A project to evaluate current practices in fine art image reproduction is conducted in which pieces of artwork in various media are being imaged by participating museums. As part of this project, observers were asked to make adjustments in softcopy fine art reproductions. The goal is to see how people working in museums, libraries and archives make color adjustments to artwork presented to them on screen. Observers were led through various interfaces allowing them to adjust the image seen on the screen to better represent the original in a light booth. They were asked to make adjustments until the screen image was 'good enough,' as an exact match may neither be possible nor necessary for us to detect relevant patterns in adjustment among observers. Patterns or trends in adjustments by observers can be used as an indication of how images should be processed to match with the adjustment by observers most closely. The adjustments by observers were compared with the prediction by three chromatic adaptation models. Overall the Fairchild92 model outperforms the Bradford and CAT02 transformations in matching with adjustments by observers more closely. Introduction The use of soft proofing in cross-media color reproductions is becoming more and more popular with the wide availability of display and computing technologies. In museums, visual editing and retouching of digital images of the collections are performed by experts for online access by visitors and researchers in the form of catalogs and postcards, for example. The demand to have the softcopy match the appearance of the original hardcopy closely is apparent. To achieve the goal, not only do we need to understand how images of artwork are visually edited, but we also need to learn the difference in perceiving color on self-luminous objects (e.g., display) and hardcopy surfaces (e.g., paper). The use of soft proofing in the hardcopy workflow is not yet well established in the museum world. This experiment will also inform us on what needs to be done to build acceptance for the use of soft proofing in this field. When looking at a piece of paper under incandescent lighting, the paper appears white. However, if the chromaticity of the paper under the incandescent light is set as the display white point, we will have a difficult time seeing the display color as white. As for hard copies we are more likely to discount the illuminant color, while for self-luminous displays we hardly fully adapt to the white point, if it is further away from that of the natural daylight. ‘Discounting the illuminant’ refers to the cognitive ability of observers to interpret the colors of objects based on the illuminated environment in which they are viewed. While cognitive mechanism relies on the observers’ knowledge of the illuminant, thus being inactive when viewing softcopy, sensory mechanism is always active, as it automatically responds to the stimulus energy. Modern chromatic adaptation models are able to predict appearance matches across different media by accounting for incomplete chromatic adaptation. As a result, cross-media color reproduction is facilitated by using such calculations to predict color matches across different media and illumination conditions. In the experiment, observers were asked to adjust images off the camera to match with the original artwork in a light booth. Three chromatic adaptation transforms (CATs), Bradford, Fairchild92 and CAT02 6 were selected to predict adjustments by observers. Bradford transformation is essentially a von Kries transformation with an additional exponential nonlinearity on the blue channel. In the experiment, the linearized Bradford transformation is included, given that it is the default chromatic adaptation in the latest ICC profile specification (ICC Version 4.2.0.0). The simplified Bradford transformation does not account for incomplete adaptation, while the CAT02 and Fairchild92 models are linear in nature and both can predict incomplete adaptation. Another distinction is that the Bradford and CAT02 models transform from tristimulus values to a ‘spectrally sharpened’ cone space while the Fairchild92 model converts to cone response directly by the Hunt-Pointer-Estevez (HPE) matrix. The von Kries predictions obtained using sharpened responsivities tend to be more color constant than von Kries predictions obtained using cone responsivities. However, negative responsivity at some wavelengths are found in ‘spectrally sharpened’ cone space, thus making it physiologically implausible. It is still under debate whether HPE or CAT02 matrices yield more accurate prediction for chromatic adaptaion. Experiments Observers were asked to adjust the softcopy on the display to match with the original in the light booth. Because it was of interest to understand how experts visually edit images, observers were allowed to make adjustments rather than having them provide data for static stimuli. The experimental setup is shown in Figure 1. A 30” Apple Cinema Display was used for showing softcopy reproductions, and an LMT 1210 colorimeter was used to characterize the display. The display characterization model proposed and detailed by Day, Taplin and Berns was followed to ensure accurate mappings between LCD digital counts and XYZ tristimulus values. Display white point and luminance were adjusted to match with those of the light booth by using a Halon perfect reflecting diffuser (PRD). Additionally, the luminance and chromaticity of the background of the light booth were measured using a PhotoReserach-650 spectroradiometer. The background of the software interface was adjusted to match these settings. The colorimetric performance of the display was evaluated. The mean and max color differences were 0.62 and 1.42, respectively.