@inproceedings{Berns2011_2,<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Abstract = {Museum lighting design is a balance between preventative conservation, curatorial aesthetics, architectural constraints, and adhering to the artist’s intent (when known). For works of art that are highly susceptible to damage from optical radiation, tungsten lighting best meets this balance. With the introduction of solid-state lighting and novel filters and lamp design for tungsten, there are more choices of correlated color temperature and spectral power distribution. Both impact a work’s appearance and there are two approaches to evaluate this inter-relationship. One is to observe the work under the proposed lighting condition, sometimes not possible. The second approach is to image the work such that its spectral reflectance is obtained at a sufficient spatial resolution for visualization. Using colorimetry and a chromatic adaptation model (CAT), the work can be rendered for a given test light. Unfortunately, very few works of art have been imaged spectrally; most are either color managed or simply unassigned RGB. If we make assumptions about likely pigments used by the artist, it is possible to estimate a work’s spectral properties, enabling lighting comparisons using color-appearance models. Because these computations require non-linear optimization, such analyses are impractical, especially for lighting designers.
The approach used in this research was to use the computational tools that are contained within Photoshop’s color management “color engine.” The first step was to build an optical database of artist materials to enable determining the relationship between a set of colorants and their spectral color gamut. Next, at least three chromatic paints plus white were selected representing the work’s palette. Using optimization, a multi-dimensional look-up table (MLUT) was created relating CIELAB and concentration. A second MLUT was produced relating concentration and CIELAB for any illuminant and observer of choice (including a CAT). The two MLUTS were concatenated and used to produce an ICC abstract profile (LAB to LAB).
An acrylic-dispersion paint palette was defined containing cadmium yellow medium, phthalocyanine green (blue shade), cobalt blue, quinacridone magenta, and titanium white. An abstract painting was created with these paints and imaged spectrally. Comparisons between the painting’s measured color under illuminant A and its prediction using the profile verified the effectiveness of this approach.
Such profiles, particularly when created for specific palettes (e.g., old master, impressionism, etc.) can be used as a tool for museum lighting design.},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Address = {},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Author = {Roy S. Berns and Farhad M.  Abed},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Booktitle = {, },<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Keywords = {color},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Month = {June},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Number = {},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Organization = {},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Pages = {},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Title = {Photoshop as a tool for museum lighting design using apriori colorant optical data},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Url = {http://www.aic2011.org/downloads/AIC_2011_Book-of-Abstracts_web.pdf},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Volume = {},<BR>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;Year = {2011}