Full text: Reports and invited papers (Part 4)

After linear conformal transformation of all the points of each stereo 
model, the transformed co-ordinates of all points common to the neigh- 
bouring models are averaged (P1 from 4 models, P2 from 2 models, P. 
from 3 models, etc.). A 
Next, the residuals (in points P. and P ,., for instance) have to be 
distributed over all points (Q, e.g.) of the model. This is done by applying 
an affine transformation to each triangle (C, P. P, ,,) formed by two 
successive edge points and either the point C, (centre of gravity of the 
upper half of the model, and kept in its place) or C, (lower half). In this 
way the relative accuracy within each triangle and, thus, within the stereo 
model, is maintained while the neighbouring models are made to fit each 
other exactly at the common points. Sander's method is suited to compu- 
terisation. 
Since 1972 on the other hand, a more refined—and more general— 
method of interpolation has become available for application to photogram- 
metric problems: the method of linear least squares interpolation (ref. 
Kraus and Mikhail, Photogrammetric Engineering 1972: Linear Least-Squares 
Interpolation). Starting with the same averaged co-ordinates of the edge 
points (P, , P,, etc.) as in the Sander method, application of the method of 
least squares interpolation will result in a more homogeneous (non-linear) 
transformation of all points of the stereo model so that the best possible 
fit to the frame of averaged co-ordinates is obtained. This fitting is funda- 
mentally better than that of Sanders—who does not leave any discrepancies— 
because the method of linear least squares interpolation automatically 
distinguishes between systematic and accidental error components and only 
corrects for the former. 
The Reichenbach photography was taken in 1959, using non-polyester 
negative film material. Since the introduction of polyester film, however, 
significantly better results have been obtained than those of the Reichen- 
bach experiment. 
1.4.1.2 Numerical Restitution of Cadastral Boundaries formed by 
Topographic Features 
In many countries, the cadastral boundaries are formed by natural or 
man-made features such as hedges, ditches, fences, walls etc. Experience has 
 
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.