und
is
age
cient.
“hina
id 120
alysed.
nation,
aph is
th the
imation
spans
ass of
ves of
Fig.1
greater
e along
and in
y. The
ement.
rrected
gramme-
ger the
owever,
larger
ts. We
g-force
n that
as the
rve of
smaller
dicates
part of
nds.
56.317 um
- Length change along x
56.29r
56.277
Se zs NO. of film
. + 1 4 6 A “3 i L 12
56.91r
Length change along y
56.89}
56.87F : ;
Fig.2 Change of film 120
2.2 Determination of Internal Orientation (Inclu-
ding the Lens Distortion) of Non-Metric Camera
The determination is aimed at a research for the
change of the internal orientation (including the
lens distortion) of non-metric camera in all
course of photography. For this purpose, the same
external geometric condition should be kept in
photography. Thus, the camera is fixed as far as
possible; The 3-D control field is repeatedly
photographed; the points with same number and
location in each photograph are used (19 points
every picture) and image coordinates measured by
same operator. The season for adopting these steps
is due to the fact that the influences of changes
of internal and external orientation for camera
are correlated each other. The application of same
photographic condition could be deemed that the
effect of change of external orientation is
basically same for each film. Equally, the
selection of similar number and location of
control points and the observation by same
operator could be thought that the effects of the
distribution of control points and the
jdentification and measurement errors of picture
points are primarily similar for each film too.
Thus it is good for us to compare the relative
variation of parameters (including the lens
distortion coefficients) of non-metric camera
computed by each film and appraise their stability
in the process of photography of all film roll.
The film are measured on the large tool microscope
which can read by estimate the coordinate up to
0.1um. Then the solution is done with following
procedure
(1) correction of film deformation
xsx'(dx'/dx), y=y'(dy'/dy)
Where x',y', measured image coordinates; X,Y,
corrected image coordinates; dx,dy, frame spans
measured in film along the lines of x and y; dx’,
dy*, corresponding practical frame spans of
camera.
(2) The internal and the external orientation ‚of
camera are calculated by the space resection with
a single photograph:
ai (X-Xs)+b1 (Y-Ys)+c1(Z-Zs)
x-x,+Ax= f
a2 (X-Xs) *b2(Y-Ys) *c2(Z-Zs)
a3 (X-Xs) *b3 (Y-Ys) *c3 (Z-Zs)
Y-Y. tAy- Î
a2(X-Xs) *b2(Y-Ys) *c2(Z-Zs)
where axsk1(x-x.)r?,ys-kl(y-y.)r?,r?z(x-x.)?*t(y-y. )?
Every symbol in equation is known to all.
The photographs are measured and calculated about
three intervals. The computational results are
listed in table 1 and 2 separately
Table 2 Principal distance and lens distortion of
"seagull"4D
NO. of film i. 4 6 9 12
f (mm) 75.820 375.993 75.831 75.879. 75.850
k1-1076 4.730 5.480 5.938 6.113 5.835
residual(mm) 0.016 0.014 0.014 0.014 0.013
Note * The residual of point means the standard
error estimated by the coordinate residuals of
image points in space resection computation
according to equation m-t/my7*my?
** The coordinates of principal point is not
listed in table because there aren't collimation
marks in camera and a known point is used to be
origin in computation.
2.3 Discussions
2.3.1 The effect of lens distortion in non-
metric camera: Based on the results calibrated
above, The effect of lens distortion of two non-
metric camera made in China is shown in table 3,
in which the data of a Japanese camera is also
given for comparison.
Table 3 The Mean Calibrated Data and The Biggest
Influence of lens Distortion of Three Camera
Model formal Princ. Pho. Distor. biggest dr
of dis. dis. coef. influence
camera (mmemm) (mm) (m) (k1) (mm)
MAMIYA C330 36-24 55 6 1.2:107°* 0.095
(Japan)
Seagull DF-1 36-24 58 6 .1.7-10% * 0.292
(China)
Seagull 4D 55:55 75 6 5.6-10 ^ 0.208
(China)
*The data is supplied by photogrammetric Depart-
ment of Wuhan Technical University Of Surveying
And Mapping
Table 1 Principal distance and lens distortion of "seagull" DF-1
NO. of film 4 7 11 . 14 20 22. 24 23) .30 .32 36
f (mm) 58.765 .786 .843 .853 .801 .811 .840 .802 .840 .795 .832
K1:1075 3.510 .958 .728 .653 ‚675 .845 .774 .557 .669 .668 .791
residual(mm) 0.007 .007 .008 .006 .008
.007 .008 .008 .005 .007 .008 .009