film 
lours 
pility 
urs 
rhtly 
>rformance 
In ten small test areas a (nearly) simultaneous ground count was carried 
out. 
The data take-off was done with transparencies, copied from the negative 
film, on a light-table with an Old Delft scanning stereoscope, 43 times 
magnification. 
Counting of the parked vehicles by an inexperienced interpreter, and 
plotting their positions on maps, scale 1:5000, took slightly more than 
40 hours for the 30 km2 urban area. About 500 vehicles/km2 were parked 
on public streets. A differentiation was made between cars and trucks. 
Exposure conditions and the scale of the photographs were near the limits 
of car-detection capability. The filtering proved to be critical: the 
minus-blue anti-haze filter created some detection problems (particularly 
dark blue and green and black cars in shadow areas). 
The comparison with the ground survey revealed a probable under-estimation 
of about 5% (or less) in areas with high buildings (screening vehicles), 
shadows or trees (uncertain detection of some cars). An exact comparison 
is not possible, as the field count took more time than the overflight and 
was therefore not fully synchronized; moreover, some problems of definition 
of public street occurred in the field survey, which could not be adjusted 
afterwards. 
It was clear afterwards that the area could have been flown at a larger 
scale (with matching photo-interpretation advantages) before the specified 
time limit,with the very good flight performance realized. The usual safety 
margin for navigational errors, turns, etc. would then have to be narrowed, 
which gives more chance for not meeting the time specifications. 
The use of a minus-blue filter seems to be questionable in this special 
case; the trade-off between haze elimination (moreover exposure factor 2x) 
and detection capability of some dark-colored vehicles should be established 
by further research. 
Color film is normally very advantageous for parking and traffic aerial 
surveys, but in this case the higher Effective Aerial Film speed of black- 
and-white film was decisive. 
General conclusions about the efficiency of the aerial survey as compared 
to à ground survey will not be given here, as the data provided above do 
not permit drawing such general conclusions and local relative cost of 
aerial and ground surveys will vary considerably. The major advantages 
of an aerial survey, the high temporal correlation between data, the multi- 
purpose data source character and the permanency of the record (checking 
afterwards possible) may be,but are not invariably,decisive in specific 
local circumstances.