A lively discussion followed the Chairman’s remark that he considered it unimportant to elaborate on the 
various interpolation procedures used in DTMs, since the trend seemed to be towards linear interpolation 
methods - the modern registration equipment having reduced the efforts involved in measuring large numbers 
of points. Although the conclusion was not disputed, it was argued that it is the nature of the terrain which 
favours linear interpolation methods, a result which has been confirmed by both theoretical and experimental 
studies. 
A number of systems were then briefly described. In the U.K., the BIPS system, initially based on a square grid, 
has now been replaced by a string DTM, consisting of digitized contours and breaklines. Mexico uses the 
French system of applying 2nd degree surface fitting functions to an irregular grid. They are at present 
attempting to include breaklines in the system. In Germany, the main aim was to reduce the processing time, 
and, therefore, the two phases of data collection and calculation have been optimized independently, various 
input forms such as regular or irregular grids and digitized contours being first transformed to a regular grid 
pattern before commencing with the linear interpolation. In the Netherlands, the necessity of producing 
additional profiles has been overcome by intersecting the available digitized line detail with vertical planes for 
each profile required. The variable grid is favoured in Norway, since this permits the storage of less data in 
uncomplicated terrain forms. In Finland, an interpolation system has been developed which uses the technique 
of matrix arrays. The system allows the efficient calculation of volumes, areas, etc. from points digitized in 
various co-ordinate systems such as rectangular, polar, cylindrical and spherical. Finally, mention was made 
of the Wild digital automatic contouring system which produces both a regular grid D.T.M., and digital 
contours, from completely irregular input data. 
The problems involved in checking the validity and the completeness of data stored is tackled by the U.S. Forest 
Service by producing, as an intermediate product, a contour plan which will reveal any irregularities in the 
data. In the Ontario Highway Department, checks and double checks are made of all phases. This not only 
includes checking the input data by producing a plot on the internal tracing table of the Planimat, but also 
involves a series of editing checks in the processing phase. Graphical plots are produced in the U.S. Bureau 
of Public Roads, although they rely heavily on their operators to locate blunders. Finally, mention was made 
of the checking facility now available with the graphical display system connected to the Konigsberg automatic 
plotter. 
As far as problem areas are concerned, it was agreed that other applications would have to be investigated, such 
as optimization techniques for alignment and gradients, etc.; but it was also suggested that terrain classification 
methods would have to be developed for the purpose of determining cost and accuracy models of different 
D.T.M. systems. 
In summarizing this first part of the discussion, the Chairman remarked that the progress since the previous 
Congress had not been all that imposing, and he therefore considered it useful to continue the work in this 
field. 
The discussion then turned to the general application of photogrammetry in highway engineering, where the 
experience in the U.S. is that, although photogrammetry is applied in all four phases, namely route planning, 
design, right of way and construction, it is so integrated with other operations, and in particular field surveying, 
that it must only be considered as a means to an end. The trend in Italy is to simplify the data collection phase 
by connecting photogrammetric equipment on-line and in time sharing to computers. In this way, the 
recording phase is eliminated, and immediate checks of the data are possible. As far as complete systems are 
concerned, they consider that a fully automated system will not be flexible enough, and therefore propose the 
development of simpler systems where the designer has the possibility of intervening during the processing 
phase. 
This suggestion was supported by a number of speakers, the reasons being not only that very few countries can 
afford expensive systems, but also in view of the fact that account has to be taken of a large number of other 
parameters, such as land acquisition, environment, geology, etc., which affect the routing of roads, and which 
are difficult to automate. Finally, a brief description was given of a simple system used in Mexico, the route 
location being carried out using stereoscopes and parallax bars, and the preliminary design on maps produced 
on Balplex plotters. 
In summing up, the Chairman concluded that a lot of work has still to be done in both fields, and read the 
Resolution proposing that the activities in this field be continued. An omission to read the title of a presented 
paper in the previous session was also rectified, this being “The Use of Modern Methods in Highway Design 
in Sweden during the years 1968-72’ by C.O. Ternryd and B. Adolfsson.