Publisher’s version / Version de l'éditeur:
Lighting Design + Application, 39, Oct 10, pp. 18-20, 22, 2009-10-01
READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright
Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la
première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.
Questions? Contact the NRC Publications Archive team at
PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.
NRC Publications Archive
Archives des publications du CNRC
This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.
Access and use of this website and the material on it are subject to the Terms and Conditions set forth at
Research Matters: HDR making strides
Veitch, J. A.
https://publications-cnrc.canada.ca/fra/droits
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
NRC Publications Record / Notice d'Archives des publications de CNRC:
https://nrc-publications.canada.ca/eng/view/object/?id=603519d3-f6fe-4944-abbe-8fb46b20b02e https://publications-cnrc.canada.ca/fra/voir/objet/?id=603519d3-f6fe-4944-abbe-8fb46b20b02ehttp://www.nrc-cnrc.gc.ca/irc
Re se a rc h M a t t e rs: H DR m a k ing st ride s
N R C C - 5 2 6 6 3
V e i t c h , J . A .
O c t o b e r 2 0 0 9
A version of this document is published in / Une version de ce document se trouve dans:
Lighting Design + Application
,
39, (10), October 2009, pp. 18-20, 22
The material in this document is covered by the provisions of the Copyright Act, by Canadian laws, policies, regulations and international agreements. Such provisions serve to identify the information source and, in specific instances, to prohibit reproduction of materials without written permission. For more information visit http://laws.justice.gc.ca/en/showtdm/cs/C-42
Les renseignements dans ce document sont protégés par la Loi sur le droit d'auteur, par les lois, les politiques et les règlements du Canada et des accords internationaux. Ces dispositions permettent d'identifier la source de l'information et, dans certains cas, d'interdire la copie de documents sans permission écrite. Pour obtenir de plus amples renseignements : http://lois.justice.gc.ca/fr/showtdm/cs/C-42
Originally published in Lighting Design + Application, October 2009 (Vol. 39, No. 10), pp. 18-20, 22.
RESEARCH MATTERS: HDR MAKING STRIDES
By Jennifer A. Veitch
Communication between designer and client is a critical component of the design process, and still images have been part of the process for decades. Successful projects rest on realistic portrayals of scenes, whether they are photographs of real places or renderings of places yet to be built.
HIGH-DYNAMIC-RANGE IMAGING
Real places, particularly if daylit, have large luminance variations, from shadows with less than 1 cd/m2 to sunlit shiny surfaces exceeding 100,000 cd/m2. When adapted in the photopic range, the human eye can detect luminance levels over a range of 8 log units 1 and with 106 levels of resolution 2. Conventional image capture and display technologies, however, do not match the resolution capability of the human visual system. A conventional liquid-crystal display (LCD) monitor can present luminance levels up to ~ 300 cd/m2, with only 256 distinct luminance levels. Fortunately, technologies exist to increase the dynamic range for both image capture and display.
High-dynamic-range cameras exist for the professional market, but they are very expensive. A more affordable solution for high-dynamic-range (HDR) photography is to combine multiple images taken at varying exposure durations into one image by using software designed for this purpose, such as
Photosphere 3. The resulting HDR images can be calibrated against spot luminance measurements taken at the time of image capture to produce luminance maps 4.
Conventional displays cannot make use of the additional resolution in HDR images; however, there exists HDR display technology that can 5. In this device the uniform backlight of a conventional LCD screen is replaced by an array of white Light Emitting Diodes (LEDs). The image is shown on the foreground color LCD screen as in a conventional display, but it is backlit by a low-resolution version of the same image formed on the LED array. When these two layers are combined, with correct settings of the parameters of the display, an image is displayed with much high luminance values and contrast ratios than a
conventional image.
AS GOOD AS REAL?
In collaboration with Dr. Lorne Whitehead at the Structured Surface Physics Laboratory at the University of British Columbia, we have conducted three experiments using HDR displays. The first used an early version of the display 6. The second experiment, using a more advanced version of the HDR display, allowed a more robust test of the hypothesis that HDR images on an HDR screen would be perceived as more realistic than conventional images 7. In this experiment, 39 participants viewed six scenes in three modalities: the real scenes (locations in the building where the experiment took place), digital
photographs of the scenes displayed in conventional mode (with screen resolution and luminance of a conventional LCD display, at the time 200 cd/m2, and 256 distinct levels), and the same digital
photographs shown in HDR mode (capable of luminances as high as 4000 cd/m2 and 216 distinct luminance levels). Figure 1 shows two of the six scenes. The photograph of the gymnasium had the lowest average luminance but the greatest luminance variability (Figure 1a) and the photo of the stairwell had the highest average luminance but the lowest luminance variability (Figure 1b).
Half of the participants saw the real spaces first, and half saw the digital images first (randomized order). For each presentation (real and digital images), participants rated what they saw on four semantic differential scales: dim — bright; non-uniform — uniform; unpleasant — pleasant; glaring — not glaring. Participants then viewed the six digital image pairs again, and indicated whether the HDR or conventional image was more realistic.
Figure 1. Two of the six spaces viewed in the first experiment. Figure 1a (left panel), a gymnasium,
average luminance 39 cd/m2 and normalized standard deviation 2.6 cd/m2. Figure 1b (right panel), a stairway with windows, average luminance 458 cd/m2 and normalized standard deviation 0.96 cd/m2.
We focused our attention on the rating data from those individuals who had seen the real spaces last, because when HDR images are used for design decision-making, the real spaces will not be available to view. In this brief article I will discuss ratings from two of the six scenes.
We expected HDR images to be rated as different from conventional but not different from real spaces. Figure 2 shows the results graphically. Our hypotheses were supported for spaces with daylight, but not for those with predominantly electric light. For the gymnasium (Figure 2a), the HDR image was more dim, unpleasant, and glaring than the real space; its only difference from the conventional image was in seeming more uniform. When forced to choose, 21 participants said that the HDR image was more realistic, and 12 chose the conventional image, not a statistically significant difference. For the stairwell (Figure 2b), the HDR image and the real space did not differ in the ratings for uniformity and glare. The ratings of the HDR and conventional images were different for all four scales. In the forced-choice
procedure, 26 participants said that the HDR image was more realistic, and seven chose the conventional image; this difference is statistically significant.
Over all six scenes, it became apparent that the HDR images on the HDR monitor were rated as we had expected when the scene exploited the HDR features: Scenes with high luminance tended to be judged as more similar to the real spaces. In our set of scenes those were images of daylit spaces. This experiment demonstrated that HDR images are at least as good as conventional images for obtaining feedback about space appearance, and better than conventional images when the scenes feature relatively large areas of high luminances, such as daylit spaces.
Figure 2. Average semantic differential ratings of those participants who viewed digital images first.
Figure 2a (left panel ratings) are for the gymnasium scene. Figure 2b (right panel) ratings were for the stairway scene. 0 20 40 60 80 100
dim non-uniform unpleasant glaring bright uniform pleasant not glaring
real-i hdr-i conv-i 0 20 40 60 80 100
dim non-uniform unpleasant glaring bright uniform pleasant not glaring
real-i hdr-i conv-i
ARESEARCH TOOL
One of the benefits of using images for appearance judgments is greater control over the scene; every participant can see exactly the same image. The first experiment gave us confidence in using the HDR display as a research tool. In our next experiment, we studied the relationship between scene
characteristics (average luminance and view size) and space appearance judgments. In this experiment, participants saw and judged HDR images of the same office with 12 variations in the height and material of the panel between cubicles, and with variations in the position of the window blind. Figure 3 shows examples of two of the images — the most open and the least open of the scenes.
Figure 3. The view is that of the seated occupant in the second row of cubicles in an open-plan office.
Left panel: The least open scene, with a high panel and window blinds tilted. The average luminance was
24 cd/m2 and the view size was 0%. Right panel: The most open scene, with a low panel and window blinds open. Average luminance was 348 cd/m2 and view size was 100 % of the maximum possible.
The participants rated each scene on 8 scales. We plotted the average ratings for each image against the average luminance for that image 8 (Figure 4). Based on previous research we would expect that scenes would be judged as more pleasant when the average luminance increases; however, most previous work has not included the higher luminances that windows can deliver. One might expect that if luminance levels are high enough then judged pleasantness would plateau, and even decline.
In this study, the highest average luminances were around 350 cd/m2 and the maximum within any scene was ~2600 cd/m2. The pleasantness of the scene tended to increase with increasing average luminance, but started to level off around 175 cd/m2. Relative view size also predicted pleasantness in a non-linear way, leveling off around 70%. These relationships are consistent with other previous research 9. Other results for this experiment are available in a conference paper 8.
Figure 4. Average ratings of pleasantness plotted against average luminance (left panel) and view size
(right panel). 1.00 3.00 5.00 7.00 0 100 200 300 400 Pleasantness Average Luminance (cd/m2) 1.00 3.00 5.00 7.00 0 50 Pleasantness Relative View (%) 100
This experiment provided us with useful information about office design choices for panel heights and view, which we hope will influence office design practice. Using the HDR display, we were able to extend
our knowledge far beyond the 200 cd/m2 maximum luminance available to users of conventional monitors. It would have been impossible to conduct an experiment like this in a real space because changing the office furniture configuration took time, and during that time the daylight conditions would have changed.
AVISUALIZATION TOOL
High-dynamic-range displays are not yet commonly available, but industry is further developing the technology. Research has demonstrated that for scenes with high luminances typical of daylit spaces, HDR images are more realistic than images projected in the conventional way. In future, we expect that this modality will become a valuable tool to the design community to assist in the design process and to facilitate communication between designer and client.
REFERENCES
1. Kaiser PK. 2009 July 14. The joy of visual perception: A web book. York University <http://www.yorku.ca/eye/thejoy.htm>. Accessed 2009 July 14.
2. Boyce PR. Human factors in lighting. London: Taylor & Francis; 2003. 3. Ward G. Photosphere. Albany, CA: Anyhere Software; 2005.
4. Inanici MN. Evaluation of high dynamic range photography as a luminance data acquisition system. Lighting Research and Technology 2006;38(2):123-136.
5. Whitehead LA, Ward G, Stuerzlinger W, Seetzen H; High dynamic range display devices. U.S. patent 6,891,672. 2005.
6. Newsham GR, Seetzen H, Veitch JA, Chaudhuri A, Whitehead LA. Lighting quality evaluations using images on a high dynamic range display. Architectural Research Centers Consortium / EAAE 2002 International Conference on Architectural Research. Montreal, QC; 2002.
7. Newsham GR, Cetegen D, Veitch JA, Whitehead LA. Comparing lighting quality evaluations of real scenes with those from High Dynamic Range and conventional images. ACM Transactions in Applied Perception in press.
8. Cetegen D, Veitch JA, Newsham GR. View size and office luminance effects on employee satisfaction. Proceedings of Balkan Light 2008, Ljubljana, Slovenia, October 7-10, 2008. Maribor, Slovenia: Lighting Engineering Society of Slovenia; 2008. p 242-252.
9. Newsham GR, Brand J, Donnelly CL, Veitch JA, Aries MBC, Charles KE. Linking indoor environment conditions to job satisfaction: a field study. Building Research & Information 2009;37(2):129 - 147.
Jennifer A. Veitch, PhD, Fellow IESNA, is a senior research officer at the National Research Council Canada Institute for Research in Construction (NRC-IRC). She's best known for her research on lighting quality, and her contributions to Chapter 10, Quality of the Visual Environment, in the IESNA Lighting Handbook, and more recently for research into environmental and job satisfaction in open-plan offices. She is active in several professional associations including CIE, where she is secretary of Division 3 (Interior Environment and Lighting Design), and vice president of the Canadian National Committee. She serves on the IES Lighting Criteria committee and is an advisory member of the Quality of the Visual Environment committee.
