TS BASED ON
China
ell, Occlusion
ti-view image matching,
oses a fictitious plane in
Z image resolution). By
all images intersect with
ferent images projection
hich matching candidate
rojection ray in the grid
instant”, and uses initial
proposed in this paper is
o have reliable matching
S40 image. It separately
ce image and searches
nakes quality check to
d makes the least square
to corrected matching
d Ming (2009) introduce
integrating image and
zes the integration of the
he object space through
function of gross error
ach. Matching result in
does not combine multi-
rocess of matching. So it
1 object space or filter
hing results of multiple
simultaneously increases
tching to all images
ymetry constraint mode,
and the space coordinate
Zhang (2005) and Zhang,
ally constrained cross-
ooses the nadir-viewing
ence image, and extracts
and then searches the
ge. It comes through the
rence image) to object
>). Because it is restraint
iage, it does not adapt to
ective in the areas with
f a certain space area I$
not imaged in the reference image because of occlusion but
imaged in the other images, this area will not participate in
matching. This problem may be solved to take turns to select
each image as the reference image, but it increases the
computation. Vertical line locus (VLL) is used by image
correlator in DSR-11 mix digital photogrammetry workstation
(Zhang & Zhang, 2002). VLL utilizes “ground element” in the
object space as matching primitives, which have been used
multi-image matching (Zhang, et al, 2007; Ji, 2008). It is
uncertain because it integrates the similarity measure function
in all stereo pairs in the course of matching, which reduces the
effect from incorrect matching caused by occlusion and
repetitive texture by right matching.
Focusing on the serious occlusion problem in city images, this
paper makes full use of the advantage of multi-view image
matching, and proposes a reliable multi-view image matching
algorithm supported by the moving Z-Plane constraint (MZPC).
[t introduces geometric constraint used in the Space-Sweep
method (Collins, 1995), and makes multi-image matching
simultaneously for feature points. Based on Space-Sweep
method, the MZPC proposes the multi-image selective
matching strategy under the grey similarity constraint. This
algorithm can simultaneously carry on the matching of multi-
image feature points under “the best candidate will be matched
in the first instant” matching strategy and plane grid height
constraint.
2. MULTI-VIEW IMAGE MATCHING FOR FEATURE
POINTS UNDER THE MOVING Z-PLANE
CONSTRAINT
The multi-view image matching algorithm under the moving Z-
Plane constraint is based on the basic photogrammetry principle
of forward intersection that the corresponding points in the
different images will always intersect to the same object point
in the object space. In this paper, supposing that the position
where intersected by different image projection rays of feature
points may be the space position of the imaged feature, one new
object constraint mode to multi-view image simultaneously
matching based on feature points can be established according
to this hypothesis.
21 Constraint by the Moving Z-Plane
moving
plane
from
Z^ Vx
10
"hn
+
Figure 1. Sketch map of moving Z-Plane constraint
Suppose a fictitious plane in the object space, the direction of
this plane is vertical to the direction of vertical line in the object
Space, and the size of the plane contains the areas in the all
Mages in the object space with
BS Anl AT ). In Fig. 1, the plane is
dn max? ^ min max
divided into regular grid cells (small plane elements) by a
certam interval (2image resolution), which can be seen as a
grid of ground element or “groundels” in the scene. The largest
height value ox and the smallest height value Zi of ground
surface are determined, and moves the plane along the object
vertical line with a certain step in the range of Zui ~ 70 The
size of step is directly in ratio to the size of grid cell.
This paper extracts the feature points in all images. First, it
moves the plane to the largest height value position, and the
plane equation is Z=7_ . Using the inverse solution of
collinearity equation, it makes the point projection rays in all
image intersect with the plane, and obtains the object point
P(X,Y.2) in the plane, i-1,2,...,N, N is the total number of
projection feature points. Then it separately statistics the
number of projection rays in the each grid cell, and if
number » T in a certain cell (7 is the threshold), the position
of this grid cell will be regarded as the feature point position in
the object space. Finally, it records all the grid cells which with
number > T in the plane of this height, and considers them as
the grid cells to be matched.
In the height plane, for each grid cell to be matched, its
corresponding feature points in different images will be
simultaneously recorded, which means that the projection rays
pass through the same grid cell. This algorithm selects images
having feature points, and performs the grey correlation
matching. If the computed correlation coefficient is beyond a
certain threshold, it regards the position of grid cell as the
object feature point position, and the corresponding feature
points in different image will be the corresponding points. Then
the grid cell is evaluated with height value of the plane position,
which will not participate in the latter matching. This process is
called as grid cell in the plane matching.
Multi-view image and
orientation element
Extract feature point by
Forstner interest operator
Make sure the plane cover
area and height search range
The plane is divided by
regular grid
peescccass P
Initially matching with i
the best grid cells
Moving plane to different height
position from Z-Zm t0 Z-Zmin
Matching with the
second-best grid cells
Statistic the number of projection
ray in the each grid cell in the plane
1
Number? a half of the number
of matching image
2<Number<a half of the number
of matching image
Initial the corresponding
oints and grid height matrix
Y
te f Grey similarity constraint
Constraint by plane grid cell
height
Evaluate to grid height
matrix
Final the corresponding points
and grid height matrix
Figure 2. Flow chart of the overall algorithm
