International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
3t
N
E
Picture 02 — Shows the laser scanner points selected in the
region with big buildings
Of the 800,000 points originated from the laser scanning and
existing in the region contained in the image, approximately
10% of defining points of the borders of the buildings were
selected.
Figure 3: Shows the laser scanner points selected in the region
of small buildings.
5. DETERMINATION OF CONTROL AND
VERIFICATION POINTS
Fifteen control points and seventy verification points in form
of point image were identifiable in aero photos taken with an
aerophotogrammetric camera. Its coordinates, in geodesical
referential, were determined from observations conducted in
photogrammetric models previously oriented in the analytic
photogrammetric Zeiss Planiacomp C-100, property of the
Graduate Program in Geodetic Sciences at UFPR.
The aerophotogrammetric covering was conducted in July
2000 in a scale of 1/6000. A metric camera, WILD RC-10,
with focal nominal distance equal to 153.000mm was
employed. The models possess artificial photogrammetric
points (PUG). The coordinates were determined by bundle
adjustment aerotriangulation. This paper took into
consideration the fact that the coordinates of support points
and verification points proceeding from the reading of
photogrammetric models are exempt from error, to be taken as
the base for verification of work conducted with the integration
of laser scanner data and small format digital aerophotos.
6. TRANSFORMATIONS BETWEEN
PHOTOGRAMMETRIC AND GEODETIC SYSTEM
The vectorial file containing the contours of the buildings is
obtained through monocular digitalization of the image utilizing
the CAD MicroStation PC system (computer aided design).
Applying the mathematical transformations between the
photogrammetric and geodetic systems involved, one proceeds
to the rectification of the vector file generated in the
digitalization. The fundamental mathematical model performs
the transformation of photogrammetric coordinates (xp, yp) of
points observed in the image, for the local tri-dimensional,
Cartesian, geodetic ‚system. (AL„YL,ZL); utilizing inverse
collinear equations. More details on this transformation may be
seen in MAKAROVIC, (1973) and MITISHITA, (1997).
miXPt mo, YP*ma,€
Xp e Xo*(Zj - Zo) 4 25 EC (1)
m133P*m3XPtma34*
mi5Xp*m^5 YP* ma5€
Y, = Yo+(Z; - ee 0 32 (2)
SFA Tm gat
xf =axg+byg+e (3)
yf-cxg«d.yg«f (4)
Xp = Xf — Xo — Ar(Xf — Xo) — Adx (5)
Vp = Jf = Yo = Arf = yo) My (6
Ar=(kır? +k,r* +kzr®) (7)
Adx = P|(r* +2(xf — xo)“)+2P,(xf — xo)(yf — yo) (8)
Ady = 2P.(xf - xo) — yo) + Po? €2(r- yoy) ©
r? 2 (xf - xo) 4 Qf - oy? (10)
T : : ; ;
(xg vg)" = Coordinates in the graphic system;
(xf, y» — Coordinates in the image system;
Gp. y) — Coordinates in the photogrammetric system;
(abcde f)= Affine transformation parameters;
(k,, k5,k4) — Radial distortion parameters;
(A, P5) 2 De-centering distortion parameters;
xo, yo) = Coordinates of principal point in the image system;
J p pair ge Sy
€ = Focal length;
T SEY . x E ; a 1 .
Xo Yo Zo) = Coordinates of exposure station;
p
[Xy Y, Z7; jf = Coordinates in the local geodetic system;
mj; = Elements of rotation matrix (R(x).R(¢).R(®)).
In this paper, the origin of the image referential was utilized as
being the center of the image.
In conventional monorestitution applications with the
utilization of inverse collinear equations, the value of the
International Arc
coordinate
supported
application
buildings w
procedure t
points origi
the borders
with the co
02 and 03.
hybrid
geodetic) (x
photogramn
observed oi
mathematic
photogramn
the local
application
coordinate
neighbor in
point to b
coordinates
(Xj 2 Y pi 34
The propos:
MONOPLO
developed a
UFPR.
Figure 4
HR
MONO
7.1 Exterior Orie
Employing collit
adjustment (Lea
orientation of th:
adjustment, eight
were observed.
photogrammetric
conducted in the
System and the
obtained from the
the analytical phe
as described in |
obtained in the ad
