The method concept shown at the Fig.1 enables us
to determine shift vector S belonging to an
unmarked point on the glacier surface from
measurements data obtained for simultaneous
stereopair L1-R1 and different time stereopair L1-
R2.
Methodological solution of the task includes two
stages: a) detection of relative orientation elements
for L1-R1 aerial photos in S1XYZ basic coordinate
system and photogrammetric model construction; b)
detection of exterior orientation elements for R2
aerial photo in the same coordinate system and
shift values calculation.
Relative orientation is to be made on the basis of
complanarity equations and following analytical
conversion of the L1 and R1 aerial photos. Spatial
X,Y,Z coordinates for all the points may be
calculated by formulae:
X XB ; y 2—3—; (1)
N
I
/= 2
€ 7X
where B is the photography base; — x,;y, and
X,';y, are converted coordinates of the points for
L1 and R1 aerial photos accordingly; f - aerial
camera focal length.
At first B is considered as an approximate value
which has to be defined more precisely during the
model orientation process using control geodetic
points or cartographic data. A version using a free
model without geodetic control is also possible .At
the present time GPS-techique permits us to
determine the photography base value during flight
and photography.
Exterior orientation elements for the R2 aerial
photogram may be calculated from resection using
common stable points located out of the glacier
area from the next equations:
2 }
-(f e NER
ci fe Lie Ft SA Ne (gj
Z +AZ Z * AZ
f y 2
M a m. m
7 + AZ 7 + AZ EMG
where a,o,x are angular and AX , AY , AZ are
linear exterior orientation elements;
X», y» are the points coordinates for R; aerial photo.
Initially, conversion of the R2 aerial photo using
angular orientation elements is done. Then p, and
qs differences are calculated with the help of the
point coordinates on L1 and R2 rectificated
photograms. These differences are shift vector
components in the photo scale. The vector S is to
be determined by the formulae:
p. =X, a bach ony X A
s TdT a EAE
beh y Ji
ar te br a Aero AZ à
n'eut NR NE e
S = ps +45 -
Values ps, qs for stable points have to be equal to
zero in the frame of the accuracy limits. For
moveable points these values show in the scale of
the aerial photo shift of the point for period T
between photography moments of L1 and R2
photograms. Vector S is calculated in basic
coordinate system. For easy use it has to be
converted into oxy coordinate system on the L1
photogram with the help of angle x obtained during
the relative orientation process. Natural value of the
vector may be calculated after allowing for the
model scale.
Value S determined by this technique is the
projection of the movable point shift on the plane
XY in the spatial coordinate system. It differs from
tited surface shift on the value described by
formulae:
8S -S(1- cos9 + 7 sin 9), (4)
where 9 - angle of glacier surface tilt;
X4 - abscissa of the movable point on L1 photogram.
When the angle of glacier surface tilt is less than
15° this correction does not usually exceed 10%
from S. When it is necessary it may be calculated
after determination of tilt angle with the help of
300
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B5. Vienna 1996
spatial
surface.
It is nec
quality
the poir
their ac
position
its COC
compar
done b
R2 surf.
this surf
Cc
Photos
from re
lengthw
days t
surveyit
regards
photogr
parame
differen
a 200
dimensi
Scale.
Fig.2. |
icefall |
1:1400(
