![](bg6.png)
From this figure, one can remark that the
textured areas are corrected whereas the relatively
uniform zones.
Another statistical measure has been performed
to prove the consistency of our method which is not
only based on lightness adjustment. Table 2 gives
the average adjustment performed on the 3
perceptual components (J, C and h) for the image of
figure 12.
The values of this table demonstrate that not
only lightness is adjusted but also the chroma and
the hue even if for this latter the deviations are
small.
Table 2: Average adjustment values obtained from the
corrected image of figure 12.
Component J C h
Average adjustment 2.64 9.25 0.05
3.3 Validation
In order to validate our adaptation of s-CIECAM to
images, we have managed a psychophysical
experiments based on a forced choice paradigm.
These subjective experiments were performed on 17
images from the Kodak database. They were
performed with a panel of 15 observers which were
evaluated for the visual acuity and a normal color
vision.
The observers were only asked to choose the
image that seems to them better (more natural)
between an original and a corrected image in a blind
way. Three repetitions are made for each of the 17
pictures to see if the observer has a stable opinion.
The obtained results are presented by figure 13
which shows number of choice of corrected image
against original.
Figure 13: Diagram which show the percentage of choice
for the corrected image (1) against the original (2).
On this histogram we can see that in 75% of case the
image corrected by our model was preferred by the
observers.
The standard deviation is very weak and no
observers have been rejected because of the stable
evaluation they have given.
4 CONCLUSIONS
In this contribution a model based on
psychophysical experiments has been described. A
study of the influence of the chromaticity of the
background was realized with the same experiment.
This s-CIECAM was extended to images with a
method allowing taking into account spatial
information.
Different tests to validate our results were
presented and corrected pictures seem more
naturally than the original. Those results are very
encouraging and the future direction of this work is
its inclusion for High Dynamic Range rendering.
Finally another prospect is the study and the
integration of the temporal frequencies with digital
cinema as an application.
REFERENCES
M.D.Fairchild, “Color appearance model”. Addison-
Wesley, Massachussets (1997).
N. Moroney, M.D. Fairchild, R.W.G. Hunt, C.J Li, M.R.
Luo, and T. Newman, "The CIECAM02 color
appearance model," IS&T/SID 10th Color Imaging
Conference, Scottsdale, 23-27 (2002).
ITU-R Recommendation BT.500-10, “Methodology for
the subjective assessment of the quality of television
pictures “, ITU, Switzerland (2000).
ISO3664, “Viewing Conditions for Graphic Technology
and Photography”, ISO (1999).
M.R. Luo and R.W.G. Hunt, "The Structure of the CIE
1997 Colour Appearance Model (CIECAM97s),"
Color Res. Appl., 23, (1997).
G.M. Johnson, “The quality of appearance”, AIC 2005,
Granada Spain, (2005).
B.A. Wandell “Foundations of vision”, Sinauer
Associates, Inc, Sunderland, Massachussetts(1995).
C. Larabi and O. Tulet, “Study of the influence of
background on colour appearance of spatially
modulated patterns”, CIE, Paris(2006)
G. Sharma, W. Wu, E. N. Dalal, “The CIEDE2000 Color
difference Formula: Implementation Notes,
Supplementary Test Data, and Mathematical
Observations,” CR&A, 30(1), pp. 21-30, 2005.
M.D. Fairchild and G.M. Johnson, “The iCAM framework
for image appearance, differences, and quality,”
Journal JEI 13, 126-138 (2004).
Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli,
“Image quality assessment: From error visibility to
structural similarity”, IEEE TIP 13 pp.600-612, Apr.
2004.
USE OF SPATIAL ADAPTATION FOR IMAGE RENDERING BASED ON AN EXTENSION OF THE CIECAM02
133