ing and 
ulations 
> tiptoe 
  
  
  
acement 
12.1 cm 
current 
ie tiptop 
he aged 
ates that 
| current 
  
  
  
e tiptoe 
> table 3 
k as the 
iged and 
ride was 
teresting 
he same 
| doesn't 
s. 
Table 4. Average value of stride 
  
  
  
  
Average Stride(m) 
Flat Course Slop Course 
Aged 0.864 0.862 
Young 1.208 1.134 
  
  
  
  
  
Furthermore, acceleration and torque around a waist was 
computed using 3D coordinates which were acquired. Figure 6 
shows the relationship between the acceleration and distance of 
a tiptoe. 
It can be seen from the Figure 6 that the young have large 
acceleration and deceleration at tiptop off and heel strike 
respectively. On the other hand, the aged have low acceleration 
and high deceleration compare with acceleration. It may be said 
that the aged have not enough force for move ahead. 
5. CONCLUSION 
Gait analysis was performed for evaluation of road structure 
from the view point of universal design using HVT system, 
artificial sidewalk and the aged experience kit in this paper. 
Furthermore, image processing procedure for auto tracking, 
subject and feature extraction were investigated. 
Regarding image processing, auto tracking procedure without 
markers was achieved using the ability of the HVT, and it was 
10.00 
9.00 
8.00 
7.00 
6.00 
5.00 
4.00 
Acceleration(m/s) 
3.00 
2.00 
1.00 
0.00 
  
shown that the feature extraction without markers was 
successfully performed by mathematical morphology. 
On the other hand, it was recognized that the current regulation 
for the difference in level and slope were appropriate values. 
There are some risks of tumbling or stumbling for the aged was 
indicated even current regulation is satisfied. 
In generally, it is concluded that the HVT system was useful 
instrument for human gait analysis, in particular auto tracking 
without markers. However, there are issue which need to be 
resolved before this system may become more operational in 
gait analysis. This problems are accuracy aspects for the 
skeleton points and extraction of foreground from complicated 
background. 
Reference from book: 
Dougherty, E., 1992. An Introduction to Morphological Image 
Processing, SPIE PRESS, Washington, pp. 31-62. 
References from Other Literature: 
Anai, T. and Chikatsu, H., 2000, Dynamic Analysis of Human 
Motion Using Hybrid Video Theodolite. In: International 
Archives of Photogrammetry and Remote Sensing, Vol.X XXIII, 
Part B5/1, pp.25-29. 
Chikatsu, H. and Anai, T., 2000, Development of Hybrid Video 
Theodolite and Application for 3D Modeling of Panoramic 
Objects. In: International Archives of Photogrammetry and 
Remote Sensing, Vol.XXXIII, Part B5/1, pp.130-136. 
——$—— Young acceleration 
—lli— Aged acceleration 
-3500.000 -2500.000 -1500.000 -500.000 500.000 1500.000 2500.000 
Distance(mm) 
Figure 6. Relationship between distance and acceleration 
-273- 
  
RO Coo c CAECUS a