| = 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
  
  
  
  
  
  
  
  
  
  
  
  
  
Is) 
a 
dai + 
9 
e 
10 À 
/. 
sT » 
0 5 10 1 in + E x 50 
Frame 
(a) Low density: à = 0.1 
0 T T T T T T T T T T 
True 
3 3 sensors 
4 2 sensors 
5 6, sensors 
] p ] 
20 x 29 N 
I \ 
a / D 
= S 
315 = = 15} + 
Oo o 
o IS 
| 10 4 15 F 
0 10 i 20 2 30 40 45 5 : o 10 15 20 25 3t 4t 4 Ü 
Frame Frame 
(b) Mid density: à = 0.4 (c) High density: à = 0.8 
Figure 8: Transition of Measurement Ratio 
era and laser scanners will be developed. Improvement and face eling, Proceedings of IEEE International Conference on Com- 
identification of pedestrian will be able to be achieved. puter Vision and Pattern Recognition. 
Simon Mochon and Thomas A.McMahon, 1980, Ballistic Walk- 
REFERENCES 
ing, Journal of Biomechanics, 13, pp.49-57. 
Chia-Jung Pai and Hsiao-Rong Tyan and Yu-Ming Liang and 
Hong-Yuan Mark Liao and Sei-Wang Chen, 2004, Pedes- 
trian detection and tracking at crossroas, Pattern Recognition, 
SUC 37,pp.1025-1034 
fthe | Sp. iii 
| Huijing Zhao and Ryosuke Shibasaki, 2003, Pedestrian Tracking 
using Multiple Laser Range Scanners, Proceedings of Computers 
on Urban Planning and Urban Management. 
eier Huijing Zhao and Ryosuke Shibasaki, 2001. A Robust Method 
ene for Registering Ground-based Laser Range Images of Urban Out- 
i door Objects, Photogram metric Engineering & Remote Sensing, 
nen 67(10),pp.1143-1153. 
strian 
/iron- Naoya Ohta, Masanobu Kiyama and Kenichi Yoshikawa, 2003, 
that it Pedestrian Detection and Identification Using Two Cameras(in 
e and Japanease), /PSJ Journal, 44(10), pp.2459-2468. 
mera. 
Romer Rosales and Stan Sclaroff, 1998, Improved Tracking of 
 cam- Multple Humans with Trajectory Prediction and Occlusion Mod- 
1265