374 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1976 
TasLE B-4 (Test B-1). Test oF THE Law (EQuaTioN B-6) (r = RMS SPATIAL RESIDUAL FOR 
n CoNTrOL POINTS; 7, = MAXIMUM ACCURACY) 
  
  
Photopair n (number of control points) 
reference 7 10 15 23 28 7, 
101 r 7.21 um 7.10 6.14 5.75 5.38 um 
rir, 1.34 1.32 1.14 1.07 
102 r 7.34 um 7.15 6.83 6.57 5.78 um 
rir, 1.27 1.24 1.18 1.14 
106 1121 9.18 9.25 9.10 7.97 
1.41 1.15 1.16 1.14 
107 17.28 14.19 13.46 14.30 12.18 
1.42 1.17 1.1 1.17 
11 30.88 29.56 29.05 25.61 
1.21 1.15 1.13 
12 34.38 33.43 33.82 29.11 
1.18 1.15 1.16 
(rir,) 1.36 1.21 1.15 1.15 113 
V1 + 11/2n 1.34 1.24 1.17 1.11 1.09 
  
TasLE B-5 (Test B-2). TEsT or THE Law (EQUATION B-7) (r = RMS SPATIAL RESIDUAL FOR N 
CONTROL POINTS; r, = MAXIMAL ACCURACY) 
  
  
Photopair n (number of control points) 
reference 4 6 10 15 23 28 To 
101 r 9.03 um 7.80 8.50 7.68 7.60 7.18 um 
rir, 1.26 1.09 1.18 1.07 1.06 
102 r 9.40 um 7.98 8.73 7.90 1.73 7.38 um 
rir, 1.97 1.08 1.18 1.07 1.05 
106 11.73 11.85 9.80 9.55 10.03 9.30 
1.26 1.25 1.05 1.03 1.08 
107 19.00 22.18 16.29 15.75 18.23 16.12 
1.18 1.38 1.01 0.98 1.13 
rir, 1.21 1.20 LH 1.04 1.10 1.05 
VI + 7/3n 1.26 1.18 1.11 1.07 1.05 1.04 
  
TABLE B-6 (Test B-3). Test oF THE Law (EQUATION B-8) (p = RMS ResipuaL PARALLAX FOR n 
IMAGE-PoiNTS; p, = MINIMAL RMS PARALLAX). 
  
  
Photopair n (number of points for relative orientation) 
reference 6 9 10 15 23 28 Po 
101 p 6.18 um 5.48 5.37 5.53 5.30 4.61 um 
pip, 1.34 1.19 1.17 1.20 1.15 
102 p 7.23 um 7.13 7.08 6.15 5.93 5.53 um 
pip, 1.31 1.29 1.28 1.11 1.07 
106 6.23 6.25 5.98 5.78 
1.21 1.22 1.16 1.12 
107 6.55 5,75 5.63 5.40 5.40 
1.38 1.21 1.19 1.14 1.14 
11 13.35 13.20 13.60 12.68 12.08 10.74 
1.24 1.23 1.27 1.18 1.12 
12 13.38 13.15 13.50 13.05 13.30 11.01 
1.22 1.19 1.23 1.19 1.21 
p/p, 1.30 1.22 1.23 1.16 1.13 1.14 
V1 + 5/n 1.35 1.25 1.22 1.15 1.10 1.09 
  
in the accuracy of final determinations in space when increasing the number of image 
points. For the exception, however, the improvement was about 50 per cent. 
I explain this fact as follows: the criterion of the RMS residual parallax is not an accuracy 
criterion on the model, because slight errors in focal length values or principal points posi-