![](bg9.png)
mode adaptation was the preferred one. This may be
due to the influence of some variables, especially,
luminosity.
Further research is needed to improve the colour
adjustment algorithm. A limitation of the present
algorithm is that it darkens the symbol when the
value component of HSV is above 50%. This
threshold should be increased to preferentially light-
en the symbol. Another enhancement is to consider
the second dominant color of the surrounding back-
ground to detect similarities between the symbol and
the background.
Another extension to this study is to consider dif-
ferent lighting conditions, using a broader range of
illuminance values, including, for instance, direct
sun light exposure in a bright sunny day.
Taking into account the preferences expressed in
our previous study, we expected that the selection of
the symbols would be performed faster and more
accurately when considering adding a border adapta-
tion than adjusting colour luminosity (hypothesis
H3). Actually, the results do not show significant
differences, as they were influenced by some of the
controlled variables. We also admit that there were
few symbols to be detected due to the limited size of
the screen. The experiment could probably be im-
proved by exposing participants to a larger number
of symbols over periods of time, instead of consider-
ing s fixed number of symbols superimposing a stat-
ic background image.
This study reinforced the results obtained in our
prior work in that adding a border is preferred over
adjusting the colour luminosity (hypothesis H4) re-
gardless of the outdoor luminosity conditions.
7 CONCLUSIONS
Given the results from our previous study, leading to
the conclusion that the two favourite adaptations
were adding a border and adjusting the colour lumi-
nosity, our goal in this paper was to evaluate if these
adaptations maintained symbol’s semantics.
We investigated preferences regarding two alter-
native modes: adapting only the symbols that might
be imperceptible from the background versus adapt-
ing every symbol in the image. That is, we assessed
if the adaptation of only some of the symbols could
confuse the observer, raising the question of why
supposedly equivalent symbols look different. The
user study was performed outdoors with a mobile
handheld device in conditions close to real use.
The main findings of our study were: we con-
firmed the result obtained in our previous work that
adding a border is preferred over adjusting the col-
our luminosity regardless of the outdoor luminosity
conditions; we concluded that with border adapta-
tion all symbols should be adapted to preserve se-
mantics; and we identified also the same tendency
when colour luminosity adaptation was used.
Ongoing work explores these approaches in AR
scientific data visualization, which is particularly
demanding regarding semantics preservation, using
a tablet instead of a smartphone. Further research is
needed concerning other types of symbols and adap-
tations, and a broader range of lighting conditions.
ACKNOWLEDGEMENTS
We thank the Portuguese Foundation for Science
and Technology (FCT) and the R&D unity LabMAg
for the financial support given to this work under the
strategic project Pest OE/EEI/UI0434/2011.
REFERENCES
Azuma, R. T. (1997). A survey of augmented reality.
Presence: Teleoperators and Virtual Environments,
6(4): 355–385.
Asmare, M. H., Asirvadam, V.S., Iznita, L. (2009). Color
space selection for color image enhancement applica-
tions. In ICSAP’09: Proceedings of the International
Conference on Signal Acquisition and Processing, pp.
208-212.
Carmo, M. B., Cláudio, A. P., Ferreira, A., Afonso, A. P.,
Simplício, R. (2013). Improving Symbol Salience in
Augmented Reality. In GRAPP 2013, pp. 367-372.
Dix, A., Finlay, J., Abowd, G. Beale, R. (2004) Human-
Computer Interaction, Prentice Hall.
Gabbard, J. L., Swan, J. E., Hix, D. (2006). The effects of
text drawing styles, background textures, and natural
lighting on text legibility in outdoor augmented reali-
ty. Presence: Teleoperators and Virtual Environ-
ments, 15(1): 16-32.
Gabbard, J. L., Swan, J. E., Hix, D., Kim, S.-J., and Fitch,
G. (2007). Active text drawing styles for outdoor
augmented reality: A user-based study and design im-
plications. In VR’07: Proceedings of the Conference
Virtual Reality, pp. 35–42.
Gruber, L., Kalkofen, D., and Schmalstieg, D. (2010).
Color harmonization for augmented reality. In.
ISMAR’10: Proceedings of the 9th IEEE International
Symposium on Mixed and Augmented Reality, pp. 227-
228.
Huang, K. C. and Chiu, T. L. (2007). Visual Search Per-
formance on an LCD Monitor: Effects of Color Com-
bination of Figure and Icon Background, Shape of
Icon, and Line Width of Icon Border. Perceptual and
motor skills, 104(2): 562–574.
SymbolAdaptationAssessmentinOutdoorAugmentedReality
395