In: Paparoditis N., Pieirot-Deseilligny M.. Mallet C.. Toumaire O. (Eds), 1APRS. Vol. XXXVIII. Part ЗА - Saint-Mandé, France. September 1-3, 2010 
48 
Figure 10: Extracted hatched area in blue 
Correct marking 
detection 
Number of 
false positive 
Hatched marking 
301 /308 (98% ) 
5 
Zebra-crossing 
88 / 92 ( 96% ) 
3 
Table 1: Results of hatched marking and zebra-crossing detection 
on a 9866-images set 
promising results have been obtained from this algorithm, which 
can be easily implemented and presents a quite fast execution 
time (about 70 km per hour with a 2GHz processor). Moreover, it 
improves road lane detection when used as a preprocessing step, 
preventing road line extraction from misdetection. 
on 308 images. Table 1 summarizes our result. Hatched mark 
ing and zebra-crossing detection failed on some images because 
of the extraction on damaged markings or because of too small 
hatched areas. False positives are mainly due to reflections on 
some vehicles. 
Figure 12: Example of experimental images with algorithm re 
sult. Blue areas represent hatching lines detection and green one 
represent zebra-crossing detection. 
7 CONCLUSION AND FUTURE WORK 
In this paper, a new technique for the detection of “repeating” 
marking on images grabbed with a front side camera has been 
introduced. Some of our road marking detection methods have 
been compared to existing works and we concluded that median 
filter seems to be the best option. In the following stage, the 
characterization of connected components in frequency and spa 
tial domains allows to extract markings of interest according to 
their characteristics. More generally, this technique can be used 
to find repeating road marking patterns on a bird’s eye view. Very 
Next stage of development will consist in taking into account the 
vehicle position with respect to the road (e.g. by using a road 
segmentation algorithm) to precisely adjust our models to specific 
cases (occurring when the vehicle is turning). 
REFERENCES 
Bertozzi. M., Broggi. A. and Fascioli, A., 1998. Stereo inverse 
perspective mapping: Theory and applications. Image and Vision 
Computing 16(8). pp. 585-590. 
Fischler, M. A. and Bolles, R. C.. 1981. Random sample consen 
sus: a paradigm for model fitting with applications to image anal 
ysis and automated cartography. Commun. ACM 24(6), pp. 381 — 
395. 
Muroi. H., Shimizu, I., Raksincharoensak. R and Nagai. M.. 
2008. Pedestrian recognition by a single camera for driver assis 
tance. In: Proceedings of FISITA 2008 World Congress, Munich. 
Germany, pp. F2008-08-118. 
Perreault. S. and Hébert, P, 2007. Median filtering in constant 
time. IEEE Transactions on Image Processing 16(9), pp. 2389 - 
2394. 
Rebut, J., Bensrhair. A. and Toulminet, G., 2004. Image seg 
mentation and pattern recognition for road marking analysis. In: 
IEEE International Symposium on Industrial Electronics, Vol. 1. 
Ajaccio, France, pp. 727 - 732. 
Se. S.. 2000. Zebra-crossing detection for the partially sighted. 
In: Proceedings of the IEEE Computer Society Conference on 
Computer Vision and Pattern Recognition (CVPR'00), Vol. 2. 
Hilton Head, SC. USA, pp. 211 - 217. 
Sehestedt. S., Kodagoda, S., Alempijevic, A. and Dissanayake. 
G.. 2007. Robust lane detection in urban environments. In: 
Proceedings of IEEE/RSJ International Conference on Intelligent 
Robots and Systems (IROS), San Diego. CA. USA, pp. 123-128. 
Soheilian, B., Paparoditis, N„ Boldo. D. and Rudant, J., 2006. 3d 
zebra-crossing reconstruction from stereo rig images of a ground- 
based mobile mapping system. International Archives of Pho- 
togrammetry. Remote Sensing and Spatial Information Sciences 
36 (Part 5). pp. (on CD-ROM). 
Uddin, M. S. and Shioyama, T., 2005. Bipolarity and projec 
tive invariant-based zebra-crossing detection for the visually im 
paired. In: Proceedings of the IEEE Computer Society Confer 
ence on Computer Vision and Pattern Recognition (CVPR’05), 
Washington, DC. USA. p. 22. 
Veit. T.. Tarel, J.-P, Nicolle, P. and Charbonnier, P, 2008. 
Evaluation of road marking feature extraction. In: Pro 
ceedings of 11th IEEE Conference on Intelligent Trans 
portation Systems (ITSC‘08), Beijing, China, pp. 174-181. 
http://perso.lcpc.fr/tarel.jean-philippe/publis/itsc08.html.