Download THE USE OF INFRARED RADIATION FOR THERMAL

APPLIED PHYSICS – OPTICS – LASERS
THE USE OF INFRARED RADIATION FOR THERMAL SIGNATURES
DETERMINATION OF GROUND TARGETS*
C. PLESA, D. ŢURCANU, V. BODOC
Military Equipment and Technologies Research Agency
Received December 21, 2004
The choice of optimal spectral band for thermal signature determination is
generally determined by a consideration of threat signature characteristics and
anticipated clutter conditions as well as scenario aspects. In this mode, the studies
show us that when we will desire ground target detection, the 3- to 5 µm band
provides better signal-to-noise ratios than the 8- to 12µm band.
This study presents some analysis of thermal signatures of ground targets
obtained in 3- to 5 µm and 8- to 12 µm bands. There are also shown the advantages
and the disadvantages of each band of the two above mentioned.
Key words: Infrared, signature, thermal, radiance, target.
1. INTRODUCTION
Infrared source can be characterized as either area sources or point sources.
Conventional infrared sources generally behave as area sources.
If you consider a source of infrared radiation, irradiance E(θ,φ) directly
incident on a sensor receiver will be given by
E ( θ, ϕ) =
I ( θ, ϕ) τ
R2
where I(θ,φ) is the source radiant intensity in watts per steradian, τ represents the
transmission of the intervening atmosphere, and (θ,φ) are the angular coordinates
of the receiver.
Ground targets generate characteristic emissions in the optical bands that are
inadvertent to their propulsion and vital to the detection process. The most
prominent of these are associated with the combustion of fuel during boost and
sustain phases [1]. Discrete frequency emissions from rotational and vibrational
transitions of water vapor and carbon dioxide molecules account for much of the
exhaust emission.
*
Paper presented at the 5th International Balkan Workshop on Applied Physics, 5–7 July 2004,
Constanţa, Romania.
Rom. Journ. Phys., Vol. 51, Nos. 1–2, P. 63–72, Bucharest, 2006
64
C. Plesa, D. Turcanu, V. Bodoc
2
The intensity of plume emissions varies with many factors such as angle of
the target relative to the detector, altitude and velocity of the target, and so on.
2. GROUND VEHICLE AND EQUIPMENT SIGNATURES
The term ground vehicles and equipment encompasses the mobile tactical
equipment employed by military forces engaged in ground combat. It includes
trucks, tanks, self-propelled field and air defense artillery, command and
communications equipment, and portable electric power generators.
Sensors that pose threats to this equipment fall into the category of visual,
image intensifier equipped, television, infrared linescan mappers, and, more
recently, FLIRs, imaging seekers, and terminally guided submunitions (including
sensor fuzed weapons).
Visual, image intensified, and television sensors are dependent on ambient
illumination for signature generation. They depend both on a reflectance difference
between the target and the background to create contrast and on the availability of
sufficient reflected ambient illumination to create an adequate signal level.
Given adequate illumination, visible and near-infrared signatures ultimately
depend on the spectral reflectivity differences between the target and the
background in the sensor response band. Visual sensor can use photopic color
differences as a discriminant. Image intensifiers extend the visual spectrum out to
approximately 0.9µm or into the near infrared. So do television sensors than can
use silicon detectors with response out to approximately 1.1µm. These nearinfrared sensors can exploit the high reflectivity of live foliage and the low
reflectivity of conventional paints to see a large negative contrast difference
between the vehicle and its background.
FLIRs and imaging seekers nominally work in the 3- to 5- or 8- to 12-µm
bands. In the 3-5µm band, the sun is still a significant contributor of reflected
radiation. These sensors are capable of seeing internal target detail with
temperature differences less than a degree Celsius. However, they detect target al
long ranges by seeing their hot-spot emissions. Self-emissions offer the possibility
for long-range detection and standoff attack, day or night, by human-assisted
infrared imaging equipped systems or by autonomous munitions. This fact give
impetus to understanding ground vehicle infrared signature generation mechanisms.
Each of the equipment items, if not protected, emits a set of signatures
because of its design and configuration. Although this set of signatures is unique to
the equipment type, each of signatures can be described generically to assist in
devising protective techniques.
A first component of the thermal signature is that caused by internally
generated. Engine exhaust gases are led through a muffler system to open air. In all
cases there is a resulting exhaust gas “plume” whose size and temperature varies
with the size of the engine. In most cases the muffler system is exposed to the air
3
Infrared radiation for thermal signatures
65
and is in itself a detectable signature. Radiators by their nature are exposed to the
air and thus also present detectable signatures, although not of the magnitude of the
engine or exhaust.
Most tacked vehicle and many communications systems are equipped with
small power units auxiliary to the main engine, to permit low-power operation of
communications equipment. Such auxiliary power units do generate thermal
signature, but they are of concern primarily at night, when all other elements of the
system are quiet and cool.
A second component of the thermal signature is that caused by exposure to
the sun. The effects are solar heat loading and diurnal variations. The solar heating
phenomenon begins with the fact that most mobile tactical equipment is first, made
of metal, and second, is dark in color for camouflage reasons. As a result, when
such equipment is exposed to the sun, it absorbs heat quickly and retains the heat
throughout exposure. The speed and degree of heating are directly related to the
construction of the specific equipment.
A third component of the overall thermal signature of a military unit is
influence of equipment on the adjacent ground and air. Ground tracks, exhaust
emissions, and dust clouds are the major considerations.
As the mobile equipment items transit their area of operations, wheels and
tracks impinge upon the ground and disturb the ground surface. This action results
in a heated ground track, which can be detected by thermal sensors after the
passage of the equipment, in addition to its availability as a classic visual cue to
military activity.
When the transit is made under dry condition, it is also common that the
movement action generates dust, which is thrown up into the exhaust cloud and
floats with it. Depending on air temperature and wind conditions, this exhaust
gas/dust cloud can linger in the area and present a thermal signature after passage
of the equipment.
Thermal signature almost always results from the difference, or contrast,
between the target and its immediate background. Imaging sensors see internal
target detail and external shape detail. Therefore, target signatures are defined by
their pattern features. Those features are unique only to the extent that their
proprieties differ from those in the background. Thus, resolved target signatures
depend on background intensity mean values as well as on clutter intensity
variations on a size scale comparable to internal target detail. Background spatial,
spectral, and intensity characteristics are key to target signature generation and
signature suppression.
3. PROPAGATION IR RADIATION THROUGH THE ATMOSPHERE
The utility of a particular emission line or band for determination of IR targets
signature depends on its transmission through the atmosphere, among other factors.
66
C. Plesa, D. Turcanu, V. Bodoc
4
Where path lengths are moderate and homogeneity of the atmosphere can be
assumed, it is possible to use a Beer’s law estimate to approximate atmospheric
effects.
Degraded atmospheric conditions can change these extinction coefficients
dramatically. Table 1 [1] shows some coefficients for the 8- to 12- µm band under
less than ideal weather conditions.
Table 1
Extinction Coefficient
Weather Condition
Haze
Extinction Coefficient
0.105
Light fog
1.9
Moderate fog
3.5
Heavy fog
9.2
Light rain
0.36
Moderate rain
0.69
Heavy rain
1.39
Light snow
Moderate snow
0.51
2.8
Heavy snow
9.2
Very clear and dry
Clear
0.05
0.08
An empirical expression [1] for atmospheric attenuation as a function of
wavelength and visible band visibility a figure normally available from
meteorological reports is given by
 −3.91  λ  − q 
τ A = exp 

 R ,
 V  0.55 

where V is the visibility and R the range, both in kilometers, and λ is the
wavelength in micrometers. The exponent q depends on the size distribution of
scattering particles; typical values are 1.6 for high visibility, 1.3 for average
conditions, and 0.585 V1/3 for low visibilities (<6 km).
The choice of spectral band should not be made on atmospheric transmission
alone. Other factors such as target size and contrast with background enter into the
considerations.
Figures 1 and 2 compare [2] the signal to noise ratio for two different bands
for different situations.
5
Infrared radiation for thermal signatures
67
Fig. 1 – SNR for man sized target1.
The first is based on a man –size target with no aerosol in the atmosphere and
short ranges. Note that the 8- to 12-µm band is batter for short ranges, but a BLIP
(background-limited performance) detector in the 3- to 5-µm band could
outperform the 8- to 12-µm system at ranges beyond 5km. The second figure is for
a small, high-temperature target at longer ranges. It is important to note that no
plume emissions are considered here, only hot blackbody radiation from a tailpipe,
for example, and that the higher clutter levels in the 8- to 12-µm band are not
considered. The 3- to 5-µm band is better under these conditions. However, with
current detector technology, the 8- to 12-µm band is still superior in a tropical
environment for all but very hot targets. Al long ranges and with hot targets the 3to 5-µm band could potentially emerge as superior with detector technology
improvements.
The effect of atmosphere on target to background contrast is generally the
primary concern, so a more careful definition of contrast is in order. Absolute
contrast at zero range, defined as the difference between target and background
radiance (or temperatures) as the target can be written
C A = NT − N B ,
where the subscripts T and B refer to target and background radiances respectively.
In the case of ground targets, the background radiance is understood to be that
coming from the atmosphere behind the plane of target for the following
discussions. The relative contrast is:
68
C. Plesa, D. Turcanu, V. Bodoc
CR =
6
NT − N B
N − NB
≈ T
NB
(1/ 2)( N T + N B )
Fig. 2 – SNR for small hot targets [1].
The effects of atmospheric attenuation and path emission on contrast depend
on which definition of contrast is involved. For relatively flat target and
background spectral radiance distributions, the band-averaged atmospheric path
transmittance τ can be applied to the in band radiances. In the case of absolute
contrast, the emission factor cancel and the contrast is reduced by the band
averaged atmospheric path transmittance factor. In the case of relative contrast, the
emission term be neglected in general. If the transmitted radiances are represented
by lowercase symbols and defined as
nT = N T τ + N ae nB = N B τ + N ae
where τ is the band averaged atmospheric transmittance and Nae is the atmospheric
path emission in the same spectral band, then the two transmitted contrast can be
written as
c A = nT − nB = τ( N T − N B ) = τC A
and
7
69
Infrared radiation for thermal signatures
cR =
=
nT − nB
τ( N T − N B ) + N ae − N ae
=
(1/ 2)( nT + nB ) (1/ 2) τ( N T + N B ) + N ae


NT − N B
NT + N B
= CR 

N ae / τ + ( N T + N B ) / 2
 N T + N B + 2 N ae / τ 


NB
≈ τCR 

N
N
τ
+
ae 
 B
In some cases, such as short horizontal paths, NBτ+Nae≈NB and we are left
with cR≈τCR. In these expressions lowercase symbols refer to transmitted radinaces
or contrasts, whereas symbols are zero range values.
4. SIGNATURE ESTIMATION
The radiant intensity that is emitted by any ground target (i.e., its signature) is
I=
∑ εi Lband
i
Api
(1)
where
I
= the signature (in-band radiant intensity)
εi
= the emissivity of each area element
Lbandi = the in-band radiance of each area element
Api
= the projected area of the element.
The many radiation source –plume, hot parts, skin, reflected skyshine,
reflected earthshine, reflected clouds, etc., and their temporal and spatial
variations- make an exact determination of the signature for an arbitrary target
virtually impossible.
For nontactical target it is possible to make several simplificatying
assumptions that can give a reasonable estimation of the signature for bands α and
β. The platform is assumed to radiate as a graybody. In addition, the assumption is
made that the emissivity for all elements of the target is unity. The number of
radiation sources is limited to the hot parts and body skin.
Furthermore, the temperature is assumed to be uniform over each source.
Thus, Eq. (1) can be simplified to
I = L(THP )band
∑
hot parts
Api + L(Tbody )band
∑ Api
body
where THP is the temperature of the hot parts and Tbody is the temperature of the
body skin. Usind these assumptions, analysis of several ground vehicle
measurements (Fig. 3) has shown that THP can be estimate as falling into the 283 to
70
C. Plesa, D. Turcanu, V. Bodoc
8
323-K range. The parameter Tbody is estimate to be within a few degrees of ambient
temperature (Fig. 4).
Fig. 3 – Temperature measurement.
The estimation of the pattern of the radiation produced by the hot parts can be
found [3] by using the following relationship:
y
I ( θ) = L(THP )band
( +1)
πd 2
cos d θ
4
where
d = the diameter of the engine exhaust port
y = the distance of the turbine plate from the exhaust port
θ = the angle with respect to the normal of the exhaust port.
The radiance over the band is
λ2
Lband =
∫ L( λ ) d λ
(2)
λ1
where λ1, λ2 = the wavelength limits of band. Since the band are at most 2µm in
width, one can approximate Eq. (2) by
Lband = (λ 2 − λ1 ) L(λ mid ),
where λmid=(λ2 + λ1)/2.
The radiance can be determined from the Planck function
L( λ , T ) =
C1
λ [exp(C2 / λT ) − 1]
5
9
Infrared radiation for thermal signatures
71
where C1=1.191 × 104 W cm-2 µm4 sr-1 and C2=1.438 × 104 µm K.
Fig. 4 – Temperature and humidity air on 25.03.2004.
5. SIGNATURE MEASUREMENT
A more precise estimate of the signature can be gaind from a measurement of
the vehicle. Measurement values of the platform radiation are dependent on the
conditions under which the measurement is performed; they are a strong function
of several factors, including the background, background temperature, engine
temperature, and vehicle velocity. These effects can be particularly large in the
long-wavelength band β.
Several types of radiometers, such as the Fourier transform radiometer (FTR)
and circular variable filter radiometer (CVFR) [4], can give an accurate spectral
measurement of the vehicle signature alone. The FTR uses a Michelson
interferometer, as shown as in fig.5. A collimated beam is split into two parts, each
part traveling a separate path to a reflecting mirror. The separate beams are
recombined at the beam splitter and reflected to a detector. Fringers will occur
because of interference. The amplitude of the central fringe depends on the
difference in length that each portion of the beam has traversed. For a
72
C. Plesa, D. Turcanu, V. Bodoc
10
monochromatic source, the detected amplitude varies sinusoidally as one mirror is
moved with respect to the other at a constant velocity. The amplitude of the
oscillation is dependent on the strength of the source. The frequency of the
oscillation is a function of the velocity of the mirror and the source wavelength.
For a polychromatic source, the detected voltage is a complex function of time.
The detected voltage is the sum of each frequency response caused by each
wavelength. The Fourier transform decomposes the time response into the
component frequency responses. Thus, the Fourier transform of the detector
voltage is proportional to the spectrum of source.
Fig. 5 – Block diagram of the Fourier transform radiometer.
The CVFR moves a filter wheel in front of a detector. Each position of the
wheel allows transmission about a center wavelength with a width approximately
0.05µm. Each wavelength-dependent signal is found by rotating the wheel until the
corresponding wavelength position is in front of the detector. The wheel is paused
for a small amount of time and the detected voltage is measurement. When the
response has been found for all the filters, the spectrum is complete.
Assuming the source remains constant over the measurement time, the FTR
usually can provide a higher-resolution spectrum with less radiated power and in
less time than a CVFR.
REFERENCES
1. David H. Pollock, Countermeasure Systems, Spiee PRESS, 1993.
2. G.A. Findlay and D.R. Cutten, Comparation of 3-5 and 8-12 micron IR systems, Applied Optics
28(23), 5029(19987).
3. C. Link and M. Maas, Northrop Corporation, private communication (1990).
4. F. Grum and R. Becherer, Radiometry, Optical radiation measurements, Vol. 1, Academic Press,
New York (1979).