s are shown. 
93,00 
90,00 100,00 
ues, less than 
  
Ax, 
  
  
  
  
139 
  
827 
  
  
  
esults 
  
500 3000 
esults 
Regions), we 
grade. These 
we have to 
lues. 
AK» 
  
  
  
  
  
  
  
  
  
mode 9% | 491 | 208 "NIE 
fen 36 | 224 | he 19 | 250 
median 68 ills 514 
pes 102 | eo | ne | os 862 
max is | 2205 20:7 | 3s] l 1895 
  
  
  
  
Table 9. Symmetric Relative Orientation results (Polar regions) 
  
  
0 500 1000 1500 2000 2500 3000 
^ phi 1 
—#-d phi 1 —#-d kappa 1 -lil- d omega 2 ——d phi 2 -9-d kappa 2 
Figure 10. Symmetric Relative Orientation results (Polar 
regions) 
For the Symmetric Relative Orientation, in the Polar Regions, 
we reached once more small values, less than 3/10 of grade. 
These values are bigger that the previous ones, but we have to 
underline that we worked with preliminary values and we 
explored the Polar Regipns, i.e. a very critical zone. 
8. CONCLUSION 
In this work, we meant to illustrate how to overcome a lack of 
contents in the traditional presentation of the photogrammetric 
theory, which is particularly relevant in the educational context. 
This means to contribute to form new generations of scientists, 
technicians and practisers, as well as to support the technology 
transfer, hopefully, in an international cooperation context. In 
this spirit, the authors wish to underline the relevance of both 
the peaceful use of mature and innovative technologies, and 
their utilization for a sustainable development. Indeed the 
presentation of both special and general cases for data 
acquisition in photogrammetry is becoming more and more 
important; however the presentation in non-conventional 
photogrammetry, as already said, highlight the problem of how 
to acquire the preliminary values of the unknown parameters of 
the non-linear models. 
Finally let us emphasize that, particularly in the context of 
analytical photogrammetry and, most of all, in the new context 
of digital  photogrammetry, the explanation of the 
photogrammetric concepts, via the presentation of analogue 
procedures, is obsolete. As already said, the direct derivation of 
the photogrammetric equations from the well known relations 
of 3D space geometry is easy and clear. The generality of the 
formalism solve all different problems, which are present in the 
reality; moreover the exposition of how to solve non-linear 
problems completes the presentation itself. 
215 
9. REFERENCES 
References from Journals: 
Hattori, S. and Myint, Y., 1995. Automatic Estimation of Initial 
Approximations of Parameters for Bundle Adjustment. 
Photogrammetric Engineering & Remote Sensin, 61(7), pp. 
909-915. 
Longuet-Higgins, H. C, 1981. A computer algorithm for 
reconstructing a scene from two projections. Nature, 293(10), 
pp. 133-135 
Abel-Aziz, Y. L, and Karara, H. M., 1971. Direct Linear 
Transformation into Object Space Coordinates in Close-Range 
Photogrammetry, Proceeding of Symposium on Close-Range 
Photogrammetry, pp. 1-18 
Thompson, E.H., 1959. A rational algebraic formulation of the 
problem of the relative orientation. Photogrammetric Record, 
3(14), pp. 152-159. 
Schut, G.H., 1961. On Exact Linear Equation for the 
Computation of Rotational Elements of Absolute Orientation. 
Photogrammetria. 17(1), pp. 34-37. 
Stefanovic, P., 1973. Relative Orientation — a new approach. 
ITC Journal, pp. 417-448. 
References of Books: 
Kraus, K., 1993. Photogrammetry. Diimmler, Bonn. Vol. 1, 2. 
H.-P. Pan, 1996. A Direct Closed-Form Solution to General 
Relative Orientation. Technical Report, CSSIP, Adelaide, pp. 
1-20