3. The computer time for triangulation and approximate positioning of the
Zeiss Jena strips totalled 10.0 hours. The division into 9.5 hours for triangulation
and 0.5 hours for positioning is an estimate. After a first triangulation of the strips,
a considerable number of recording errors were detected. These were eliminated and,
as a check, the triangulation of parts of the strips with a total of 79 models was re
peated. Subsequently, the triangulation of six complete strips and part of a seventh
was repeated.
Recording errors in the Nistri measurements were the reason for repetition of
the triangulation of 17 models and, subsequently, of four complete strips.
Including program read-in, the triangulation of one complete strip required from
0.4 to 0.5 hours for the Zeiss Jena measurements and about 0.4 hours for the Nistri
ТАЗ measurements. The variation in these times is caused by differences in the number
of points used for relative orientation and in the number of models. (See Table II).
4. The computer time for approximate positioning and leveling depends not
only on the number of strip transformations, but also on the number of times that the
program must be read in. Since for the transformation of most Nistri ТАЗ and Wild A7
strips control was derived from previously transformed strips, fewer transformations
could be performed here after each program read-in than for the Zeiss Jena strips.
Also, the transformations for the Wild A7 strips include repetition of the computations
after correction of the assumed river levels.
REFERENCES
[1] See various publications of the Division of Applied Physics of the National Research Council, and also:
Manual of Photogrammetry, 3rd edition, Chapter 10.4.2, published by the American Society of Photogrammetry.
[2] G. H. Schut, A Method of Block Adjustment for Horizontal Coordinates. The Canadian Survevor, Vol. XV,