The values summarized in Table 3 suggest that systematic errors
(MEZ) were small and not significant according to t-tests (p>0.1) in
comparison to the random errors (RMSEZ). Both MEZ and RMSEZ
values are in good agreement with analogical values obtained by
evaluation of photogrammetric models (Table 2). Since a perfect score
of MEZ equal to “0” does not exclude large compensating errors of
opposite signs, the spatial distribution of the residuals was inspected
to avoid false interpretation.