Information Technology Reference
In-Depth Information
3000
2500
2000
1500
1000
500
0
2
1.5
2
1.5
1
1
0.5
0.5
0
0
Fig. 14.14 Cost function of the field of view constraint in the image space
and which grows exponentially as target features move toward the image boundary.
Such a function has been proposed in [10]
e β d 2
V fv =
α
,
(14.44)
where d is the distance between the middle of the image and the object of interest,
and
are parameters to tune the effect of the avoidance.
The gradient in the image space is given by
α
and
β
V fv = 2
d x e β d 2
β
.
(14.45)
d y e β d 2
2
β
The artificial force to avoid the object beyond the boundary of the image plane is
computed by (14.28). It is transformed from the image space to the articular space
using
∂Φ
+
r
r
g fv =
Φ V fv =
( L Φ MJ q ) +
Φ V fv
,
(14.46)
q
where M and J q are the transformation matrices defined as in (14.43) and L Φ is the
interaction matrix related to the center of the image plane.
14.5.2
Simulation Results
In this set of simulations, the task is to control the position of a feature point in the
image while keeping a second feature point (the object of interest) within the field
of the view of the camera.
Figure 14.15(a) depicts the initial image from the camera. The object of interest
is very close to the border of the image. In this case, the object will move out of the
Search WWH ::




Custom Search