The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B5. Beijing 2008 
Figure 9. Scatter plot of angle of pivoting v.s. walking speed 
956 
Figure 7. Calculated probability density distribution 
Figure 8. Extracted prominently congested area 
3.4 Statistics of Walking Characteristics 
On the basis of tracking results, the relationships between 
walking speeds, angle of pivoting are derived. Figure 9 denotes 
the scatter plot for angle of pivoting v.s. walking speeds. As the 
result of Figure 9, the data have almost 2 clusters that one 
shows low walking speed but angle of pivoting has large 
variance, and the other shows the range from 0.5 to 2.0 of 
walking speed and relatively small variance. We found when 
pedestrians change their walking directions, pedestrians slow 
down their walking speed. 
0 1 2 3 4 5 
walking speed [m/sec] 
4. CONCLUSION 
In this paper, .a people tracking method by using multiple laser 
range scanners is proposed, which is relatively robust than 
cameras in congested situation. Moreover, based on the 
obtained trajectories, crowds flow characteristics such as traffic 
density, walking velocity and directionality, a degree of 
intercrossing, and those relativity are analyzed as the 
probability density by using kernel density estimation method. 
Additionally, by calculating the uniformity of trajectories’ 
directions and the traffic volume, prominently congested areas 
are extracted. Finally, we conclude our proposed method should 
be effective to analyze crowds flow characteristics on the real 
time basis even if the target space is relatively wide and 
congested situation such as railway station or shopping mall. 
REFERENCES 
Brandle, N., Bauer, D., Seer, S., 2006. Track-based Finding of 
Stopping Pedestrians - A Practical Approach for Analyzing a 
Public Infrastructure, Proc. of the IEEE ITSC, pp.l 15-120. 
Curio, C.. Edel brunner, J., Kalinke, T., Tzomakeas, C., Von 
Seelen, W., 2000. Walking pedestrian recognition, IEEE Trans. 
Intelligent Transportation Systems, vol.l, no.3, pp.l55-163. 
Demirdjian, D., Tollmar, K., Koile, K., Checka, N., Darrell, T., 
2002. Activity maps for location-aware computing. Proc. of 
IEEE Workshop on Applications of Computer Vision. 
Grimson, W.E.L., Stauffer, C., Romano, R., Lee, L., 1998. 
Using Adaptive Tracking to Classify and Monitor Activities in 
a Site, CVPR98, pp.22. 
Haritaoglu, I., Harwood, D., Davis, L. S., 2000. W4: Real-time 
surveyllance of people and their activities, IEEE Trans. PAM1, 
vol.22, no.8, pp.809-830. 
Hu, W., Xiao, X., Fu, Z., Xie, D., Tan, T., 2006. A System for 
Learning Statistical Motion Patterns, IEEE Trans. PAMI, vol.28, 
no.9, pp. 1450-1464. 
Makris, D., Ellis, T., 2005. Learning Semantic Scene Models 
from Observing Activity in Visual Surveillance, IEEE Trans. 
SMC Part B, vol.35, pp.397-408. 
Stauffer, C., Grimson, W.E.L., 2000. Leming Patterns of 
Activity Using Real-Time Tracking, IEEE Trans. PAMI, vol.22, 
no.8, pp.747-757, 
Zhao, H., Shibasaki, R., 2005. A real-time system for 
monitoring pedestrians, Proc. of IEEE Workshop on 
Applications of Computer Vision. 
Zhao, T., Nevatia, R., 2004. Tracking multiple humans in 
complex situations, IEEE Trans. PAMI, vol.26, no.9, pp.l 208- 
1221.