![](bg8.png)
ages have been described and the localization error
has been calculated a comparative evaluation has been
carried out and the computational cost has also been
studied.
On the one hand, the Fourier Signature and HOG
offer quicker results, so they constitute the most suit-
able option to do tasks in real time. On the other hand,
HOG and gist descriptors provide better accuracy in
localization tasks.
This work opens the door to new applications of
the appearance-based methods in mobile robots. As
we have shown, the appearance-based descriptors are
a suitable method to do navigation tasks. A prob-
lem is the high requirements of memory and compu-
tation time to do the database and make the neces-
sary calculations to compute the position. Once we
have tested the methods’ robustness under human ac-
tivity and changes in illumination and in the position
of some objects, the next step should be to create a
method that continuously renews the database adapt-
ing it to new lighting conditions. It can also evolve
to a system that creates more sophisticated maps to
make it possible an autonomous navigation system.
ACKNOWLEDGEMENTS
This work has been supported by the Spanish Gov-
ernment through the project DPI 2016-78361-R
(AEI/FEDER, UE): “Creaci
´
on de mapas mediante
m
´
etodos de apariencia visual para la navegaci
´
on de
robots”.
REFERENCES
Berenguer, Y., Pay
´
a, L., Peidr
´
o, A., Gil, A., and Reinoso,
O. (2016). Nearest position estimation using omni-
directional images and global appearance descriptors.
In Robot 2015: Second Iberian Robotics Conference,
pages 517–529. Springer.
Dalal, N. and Triggs, B. (2005). Histograms of oriented gra-
dients for human detection. In 2005 IEEE Computer
Society Conference on Computer Vision and Pattern
Recognition (CVPR’05), volume 1, pages 886–893
vol. 1.
Gil, A., Mozos, O. M., Ballesta, M., and Reinoso, O.
(2010). A comparative evaluation of interest point de-
tectors and local descriptors for visual slam. Machine
Vision and Applications, 21(6):905–920.
Hoang, T. V. and Tabbone, S. (2010). A geometric invari-
ant shape descriptor based on the radon, fourier, and
mellin transforms. In Pattern Recognition (ICPR),
2010 20th International Conference on, pages 2085–
2088. IEEE.
Hofmeister, M., Liebsch, M., and Zell, A. (2009). Vi-
sual self-localization for small mobile robots with
weighted gradient orientation histograms. In 40th In-
ternational Symposium on Robotics (ISR), pages 87–
91. Barcelona.
Menegatti, E., Maeda, T., and Ishiguro, H. (2004). Image-
based memory for robot navigation using properties
of omnidirectional images. Robotics and Autonomous
Systems, 47(4):251 – 267.
Murillo, A. C., Guerrero, J. J., and Sagues, C. (2007). Surf
features for efficient robot localization with omnidi-
rectional images. In Robotics and Automation, 2007
IEEE International Conference on, pages 3901–3907.
IEEE.
Oliva, A. and Torralba, A. (2001). Modeling the shape
of the scene: A holistic representation of the spatial
envelope. International journal of computer vision,
42(3):145–175.
Oliva, A. and Torralba, A. (2006). Building the gist of a
scene: The role of global image features in recogni-
tion. Progress in brain research, 155:23–36.
Pay
´
a, L., Fern
´
andez, L., Reinoso,
´
O., Gil, A., and
´
Ubeda, D. (2009). Appearance-based dense maps
creation-comparison of compression techniques with
panoramic images. In ICINCO-RA, pages 250–255.
Pay
´
a, L., Mayol, W., Cebollada, S., and Reinoso, O. (2017).
Compression of topological models and localization
using the global appearance of visual information. In
Robotics and Automation (ICRA), 2017 IEEE Interna-
tional Conference on, pages 5630–5637. IEEE.
Pay
´
a, L., Reinoso, O., Berenguer, Y., and
´
Ubeda, D. (2016).
Using omnidirectional vision to create a model of
the environment: A comparative evaluation of global-
appearance descriptors. Journal of Sensors, 2016.
Pronobis, A. and Caputo, B. (2009). COLD: COsy Lo-
calization Database. The International Journal of
Robotics Research (IJRR), 28(5):588–594.
Radon, J. (2005). 1.1
¨
uber die bestimmung von funktio-
nen durch ihre integralwerte l
¨
angs gewisser mannig-
faltigkeiten. Classic papers in modern diagnostic ra-
diology, 5:21.
Siagian, C. and Itti, L. (2009). Biologically inspired mo-
bile robot vision localization. IEEE Transactions on
Robotics, 25(4):861–873.
Valgren, C. and Lilienthal, A. J. (2010). Sift, surf & sea-
sons: Appearance-based long-term localization in out-
door environments. Robotics and Autonomous Sys-
tems, 58(2):149–156.
Evaluating the Robustness of Global Appearance Descriptors in a Visual Localization Task, under Changing Lighting Conditions
265