Contribution 
11. development of training material 
son of different control distributions 
12. development of software / Information Systems 
13. graphical presentation 
son of different measuring devices 
son of simple vs. rigid restitution 
14. test devices and procedures of Digital Photogram. 
son of different adjustment software 
Pilot Center 
son of repetitions 
cmd tha CR twa th 
compar 
compar 
compar 
compar 
6. 
son of pre- vs. self-calibration 
nd 
compar 
compar 
compar 
compar 
compar 
1. 
7. 
son of semi- vs. non-metric photography 
son of medium vs. small format 
son of normal vs. general case 
md 
2. 
8. 
> 
3. 
9. 
cmd 
4. 
10. compar 
son of minimum vs. rich control 
md 
s. 
  
Local criteria are the Mean Square Error ó; of a point i, 
the volume of the error ellipsoid, and the characteristics 
of the error ellipsoid (lengths of the axes, etc). 
2 2 2 
Ox + Oy +0, 
3 
Although the use of the local criteria is useful for each 
individual solution, for checking the accuracies in 
different areas of the object, they are of little help in 
drawing overall conclusions. Therefore additional global 
type of criteria have to be defined and computed for each 
solution. 
Such global criteria can be the RMS value corresponding 
to the mean variance D,, the Mean Standard Deviation 
D,, and the Maximum Standard Deviation D naxt 
  
43 Graphical presentations 
In order to graphically present the results, the following 
plots have been prepared for each solution: 
* Projections of the error ellipsoid of every point onto 
the X-Y, Y-Z, Z-X planes. 
* Discrepancy vector plot, showing the differences 
between the point coordinates obtained through the 
prior adjustemnts and those obtained after the S- 
transform. These discrepancies can be used to 
determine possible misfit between the individual 
dataset and the common reference frame used for the 
comparisons. 
KAIMAKA AIKTYOY 
C» 
em OO ©) 
C @ Q S Q > 
(=) e e 
Figure 4. Sample plot of the projected error ellipsoid. 
467 
S. THE RESULTS 
5.1 Concerning adjustment software 
* 8146 participants have used bundle adjustment 
software. 
19% participants have used DLT software. 
* 2596 participants have used own software. 
* Many of the programmes do not make available to the 
user (either partly or fully) covariance information. 
5.2 Concerning pre- and self-calibration 
* 48% solutions used pre-calibration. 
* 52% solutions used self-calibration. 
* In all cases the obtained accuracies are better than 
lem. 
* In the majority of the cases self-calibration gave better 
results. 
* In some cases (20%) self-calibration gave worst 
results, which means that it has not been handled 
correctly. 
5.3 Concerning semi- and non-metric photography 
* 42% solutions used semi-metric photography. 
* 53% solutions used non-metric photography. 
* 5% solutions used CCD's or camcorders. 
* Medium format, semi-metric photography gave the 
best results. 
* Small format, non-metric photography gave the less 
food results, but good enough for architectural 
emergency purposes. 
* Small format, semi-metric performed as good as the 
medium format, non-metric photography. 
5.4 Concerning medium and small format 
* 43% solutions used medium format photography. 
* 57% solutions used small format photography. 
* Medium format performed better than small format. 
* There is a strong connection between format and 
measuring device regarding the achieved accuracy. 
5.5 Concerning measuring devices 
* 62% solutions used measuring device with precision 
lim to 5 im. 
* 29% solutions used measuring device with precision 
5im to 30im. 
* 8% solutions used measuring device with precision 
larger than 30im. 
* The role of high-precision measuring device is more 
profound in semi-metric and medium format than in 
non-metric and small format. 
* The degradation of accuracy when using lower- 
precision measuring device is very high (up to 100%) 
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B5. Vienna 1996