Geography Reference
In-Depth Information
5000
5000
4500
10 MP
4500
10 MP
15 MP
15 MP
25 MP
20 MP
4000
4000
25 MP
25 MP
3500
3500
3000
3000
2500
2500
2000
2000
1500
1500
1000
1000
500
500
0
0
0
500
1000
1500 2000
Max.covered width (m)
2500
3000
3500
4000
0
10
20
30
40
50
60
GSD (cm)
(a)
(b)
Figure 8.5
Expected camera performance for various sensor sizes. a) Width of the ground footprint versus Flying height and sensor
size. b) Spatial resolution (ground sampling distance, GSD) versus flying height and sensor size for hypothetic SLR-type image
sensors of 10, 15, 20 and 25 Megapixels and focal length of 50 mm.
8.2.3.3 Motion blur
typical exposures can be as short as 1/1000 s thus yielding
a blur of 0.4 pixels. In poor lighting conditions, exposure
times can often be as high as 1/250 s in which case the
pixel blur would reach a very visible 1.76 pixels.
Motion blur can be a key driver of image quality and
can significantly offset any potential gains in resolution.
Purchasing a camera with a higher megapixel count is
not guaranteed to produce high quality imagery with an
improved resolution and potential users should think
carefully before purchasing a camera on the sole basis of
the imaging sensor size (in pixels). Digital SLR cameras
usually employ sensors with a size of 23 mm
Camera resolution should not be the only consideration
in the planning of a hyperspatial image survey. Even if
resolution and ground footprint are often the primary
concern of end users, high resolution does not guarantee
high quality. One factor that is often overlooked in the
determination of flying and image acquisition conditions
is motion blur which is caused by camera motion during
the time of exposure. The motion blur present in an image
can be calculated with the following formula:
15 mm.
This is about two thirds the size of traditional film. Whilst
camera manufacturers have dramatically increased the
pixel count for their sensors, the size of the sensor has
remained similar. Given that the sensor area remains
similar, more light will be needed in order to ensure
that each pixel on the imaging sensor receives a sufficient
sample of photons. The result is a need for increased
exposure times which increases motion blur. According
to Equation (8.2), the flying speed of the aircraft could
be reduced to offset the increased exposure. However,
significant reduction in flying speeds may not be possi-
ble since most aircraft must maintain a minimum speed
in order to remain airborne. Furthermore, reducing air-
craft speed makes the flight less stable and the associated
small-scale chaotic motion further reduces the quality
of the images. Another option would be to increase the
×
V
∗
T
∗
H
∗
f
∗
10
−
3
∗
p
∗
10
−
6
B
=
(8.2)
Where B is the motion blur in pixels, V is the velocity
of the aircraft in m/s, T is the exposure time in seconds,
H is the altitude above ground level in meters, f is the
focal length in millimeters and p is the size of 1 pixel on
the sensor in microns. Basically, this equation calculates
the number of pixels that a single point on the ground
will traverse during the time of exposure. Motion blur is
especially relevant to hyperspatial image acquisition since
the objective of the highest possible resolution requires
flights at minimal altitudes which, according to Equation
(8.2), increases the motion blur. As a typical example,
consider the case of an aircraft travelling at 40 m/s and
an altitude of 500 m. The aircraft has a 12 megapixel SLR
digital camera with a 50 mm lens. Each pixel in the sensor
has a width of 9 microns. In excellent lighting conditions,
Search WWH ::
Custom Search