of evaluating the influence of these "elevated" features on the geometric 
quality of orthophotos and contour lines, parti cularly when using automatic 
image correlation. A photograph of the test area is shown in Fig. 1 and 
a contour map in Fig. 2. 
es 
.h 
Control Data 
i 
The evaluation of the planimetric orthophoto accuracy was based 
on the X, Y coordinates of a dense net of terrain points coinciding with 
the 10 mm reseau grid in the photographs. The use of reseau crosses in 
this accuracy study has an important advantage of offering a large number 
of very well defined points distributed in a random fashion over the test 
area. Reseau points coinciding with water surfaces, buildings and trees 
were excluded fromthe analysis. In addition to the reseau points, a total 
of approximately forty distinct terrain features were selected near 
buildings and forested areas or located in deep gullies, i.e. in locations 
where more pronounced planimetric errors could be expected. 
The XYZ terrain coordinated of th reseau points and the forty 
natural terrain features were determined on a Wild A7 using the original 
photographs. The standard coordinate eC ER of the control points thus 
determined was estimated to be 0.1 m (10 um at photo scale). Asa result 
of elevation differences in the area, and camera tilts, the terrain 
coordinates of the reseau points deviate from the regular pattern of the 
camera reseau grid. The maximum radial displacement in both photographs 
was 26 m in the terrain, corresponding to 2.6 mm in the image plane. This 
displacement is an indication of the magnitude of image corrections which 
are to be introduced by the differential rectification process in this 
experiment. 
The A7 was also used for evaluation of the accuracy of height 
information submitted in the form of profile data, contours or drop 
lines. The A7 results are considered as "true" values. Plots of the 
digital profiles, contour lines and drop lines were used to correlate 
the plotted data with the A7 stereomodel. The accuracy of this evaluation 
method was estimated by ‘rep eating one of the tests and was found to be of 
the order of 0.2 m 
Test Material and Measuring Techniques Used in Evaluation of the Ortho- 
photo Accuracy 
The following orthophoto products were used in the accuracy 
study: 
1. Contact diapositive copies on dimensionally stable film from the 
original orthophoto negatives submitted by each participant. For