tep 
ling 
an, 
tep 
E & 
ure 
jer, 
emo 
aphs 
«24, 
RS, 
nic 
ter 
PRECISION IN CASE OF COMBINED POINT DETERMINATION 
Henk de Heus 
Delft University of Technology 
Department of Geodesy 
Thijsseweg 11 
2629 JA DELF T, The Netherlands 
ABSTRACT 
Land surveyors have made significant progress in the optimization of control net- 
works. For the purpose of testing and describing the precision of coordinates 
artificial covariance matrices have been developed. 
In photogrammetry too much attention is paid to the precision. However, these in- 
vestigations are mainly restricted to the photogrammetric precision influences, as 
they neglect the stochasticity of the terrestrial control points. 
In this paper the combined influence of the two on the precision of the final coor- 
dinates is described. Some ideas concerning a modified artificial covariance 
matrix for constrained control networks are included in the precision description 
of coordinates resulting from the photogrammetric block adjustment. 
1. INTRODUCTION 
In the Netherlands the adjustment, testing and quality analysis of terrestrial 
control networks is carried out in two phases: 
l. asa free network (phase I) 
2. bij connecting the free network to known points (constrained network, 
phase II). 
The main advantage of this procedure is a seperation of the statistical testing for 
outlier detection. In a free network adjustment possible (measuring) errors in the 
observations can easily be detected, because the mathematical and stochastical 
model of the free network are well defined. In this way the data-snooping can not 
be hampered by possible complicated distortions of the known points and the 
generally only roughly known variance covariance matrix of the given coordinates. 
In the second phase only the coordinates of the known points are still to be tested. 
- 329 *