Get documents about "Performance of Target Acquisition Weapons Software (TAWS) off the
Document Sample
Performance of Target Acquisition
Weapons Software (TAWS) off the
central California coast
LT Liz Scheidecker
OC3570 Project
March 24, 2005
Introduction and Background
Target Acquisition Weapons Software (TAWS), version 3.4, is a specialized
electro-optical tactical decision aid (EOTDA) designed under the Department of Defense
(DoD) to merge electro-optical (EO) systems, targets, and the atmosphere into software
for predicting EO weapon and navigation system performance. These performance
predictions are used by mission planners to make “go/no-go” decisions, to modify
mission execution tactics or weapons loads, or to evaluate the general situational
awareness of environmental conditions for Forward Looking Infrared (FLIR) imagers,
Infrared (IR) trackers, Night Vision Goggles (NVGs) and other EO systems.
To properly model a system’s performance, TAWS incorporates weather, sensor,
and mission information into each type of analysis and predicts system performance
primarily in terms of maximum detection or lock-on range. The weather information
includes humidity, temperature, wind, and aerosol analyses valid at the TAWS valid time
(Time over Target or TOT) as well as for the preceding 6 hours.
The sensor information is categorized into three regions of the spectrum: Infrared
(IR), mid wave 3-5 microns and long wave 8-12 microns; Visible including TV and NVG
systems, 0.4–0.9 microns; and Laser, 1.06 microns.
The mission information depends on the type of mission. Different types include
target acquisition/detection, target designation and tracking, close air support, helicopter,
refueling, take-off and landing, identification of pickup/drop zones, training, and search
and rescue. For a single mission, TAWS may have to include several of these tasks.
TAWS also provides particular types of sensor performance analyses:
illumination analysis, point-based analysis, and map-based analysis. These types of
2
analysis are conducted as a function of time or approach angle. For a selected time and
target location, TAWS calculates sensor information for a time series of up to 24 hours.
In addition, for a selected time, TAWS calculates sensor information for a selection of
approach angles.
Since the point-based analysis calculates detailed performance predictions for a
single location in a selected mission, the focus of this study will concentrate on a point-
based analysis of long wave IR over the Pacific Ocean. While TAWS is primarily and
consistently used by the US Air Force in their mission planning support over land, given
the proliferation of TAWS through DoD to include the US Navy, this study will focus on
its performance off the central California coast. The main purpose of this study is to
determine how well TAWS predicts the Target Contrast Model for long wave IR in
comparison to in-situ measurements of the actual ‘target’, the R/V Point Sur, and the
‘background’, the Pacific Ocean (Appendix A).
The Target Contrast Model computes inherent radiance (zero range) for up to
three “objects” in the target scene: the target, its shadow, and the immediate background.
The inherent radiance is then used to compute apparent radiance, or radiance at the
sensor’s range.
The targets are modeled using high-resolution geometries. For a particular view
direction, some of the target facets are visible to the sensor and some of them are not.
Backgrounds are modeled as a sloped plane, with the target either on the slope or at the
base of the slope.
3
The inherent radiance of the target facets, the shadow, and the background are
determined by the objects’ reflectivities and the illumination to which they are exposed.
This illumination has three components: direct, diffuse, and reflected.1
The strength of the direct illumination depends upon its angle of incidence. For
the target facets, this is computed from the target heading and the position of the sun or
the moon. For the background, this is computed from the background slope, the down
slope direction, and the position of the sun or moon. Inherently, there is no direct
illumination on the shadow.1
For the target facets, diffuse illumination and radiation reflected from the
background depend on the target heading, the background slope, and the position of the
target relative to the slope. These parameters determine how much of the sky and how
much of the background each target facet “see”. For background and the shadow, diffuse
illumination is independent of direction.1
The inherent radiance values for the target facets, shadow, and background are
computed by summing the appropriate illumination terms and multiplying the object’s
reflectivity.1
Apparent radiance values for the target, shadow, and the background at the
sensor’s range are computed from the inherent radiance. For the target, the inherent
radiance along the sensor’s line-of-sight is the weighted area average of the inherent
radiance contributions from all the visible facets.1 The background, shadow, and average
target inherent radiance values are degraded in propagating from the target scene to the
sensor. The Atmospheric Transmittance Model (ATM) computes the degradation. The
resulting apparent radiances at the sensor are used by the Sensor Performance Model
1
TAWS Version 3 Help Index, “TAWS Physical Models: IR Model/ Target Contrast Model”. August 2004.
4
(SPM) to evaluate the signature for the target-background and shadow-background
contrast pairs.2
Data and Methods
Five IR radiometers (8-14 microns) were configured on top of the bridge of the
R/V Point Sur from February 1-8, 2005, to measure the long wave IR temperature, or
radiometric temperature, of various aspects of the ship and the sea surface temperature
(SST). Three Apogee probes developed by Everest Interscience, Inc. with a 3:1 field of
view (FOV) (i.e. for every 3 meters out measures 1 meter wide aperture) were used to
take the measurements of the ship’s bow, stern, and bridge. Two Model 3600 thumb-
sized probes also developed by Everest Interscience, Inc. with a 4° FOV were used to
take the measurements of the ship’s stack and the SST. All probes took one-minute
measurements and were recorded into the CR5000 data logger developed by Campbell
Scientific, Inc.
The target locations and tracks chosen for this study were based on the different
stations where the R/V Point Sur made various oceanographic and meteorological
measurements as laid out for Leg I and Leg II of the OC3570 student cruise. Four
different target/background comparisons were made at four different locations along the
ship’s track, the first three made during Leg I and the last made during Leg II (Appendix
B).
The first comparison was taken at 36° 48.01’N, 122°12.28’W (CTD station #5)
with a target heading of 240° and TOT of 2200 UTC on February 1, 2005; the second
comparison was taken at 36° 56.30’N, 122°24.13’W (CTD station #22) with a target
heading of 240° and TOT of 1500 UTC on February 2, 2005; the third comparison was
2
TAWS Version 3 Help Index, “TAWS Physical Models: IR Model/ Target Contrast Model”. August 2004.
5
taken at 37° 07.35’N, 122°32.98’W (CTD station #37) with a target heading of 060° and
TOT of 1100 UTC on February 3, 2005; and the last comparison was taken at 36°
48.12’N, 122°12.62’W (CTD station #46) with a target heading of 240° and TOT of 2200
UTC on February 5, 2005.
In TAWS, all TOTs for each comparison agreed with the approximate time the
R/V Point Sur was conducting CTD operations and was relatively stationary. The target
heading was derived from the approximate ship’s heading it maintained to get from one
CTD station to the next along its predetermined track. All meteorological data sets were
taken from the historical data recorded from the aspirated meteorological sensors, boom
probe, and observations during the cruise.
In TAWS, the target properties consisted of all ship properties of the R/V Point
Sur which are in the TAWS database; the sortie properties (or. the aircraft sensor type)
were a generic IR sensor (1004) from 10ft above ground level (AGL) and a sensor
heading of 060° and 240°, where the sensor was either facing toward the bow or the stern
of the ship depending on the target heading; and the background properties were clear
water with ocean albedo and low clutter.
Once the model run was conducted, the results in TAWS were converted to a
tabular form with background and target radiometric temperatures listed for every hour
from 6-12 hours prior to the designated TOT to 9-12 hours following (location specific).
The target temperatures were listed as Wide-Field-of-View (WFOV) and Narrow-Field-
of-View (NFOV) in the table where WFOV is defined as the actual target detection and
NFOV is defined as the actual discrimination of the target from the background. For this
6
study, the WFOV and NFOV are the same radiometric temperatures since the comparison
was for the Target Contrast Model (zero range).
For each model run, two per location with sensor headings 060° and 240°, the
target results from TAWS were then compared to the corresponding hourly in-situ
radiometric temperature of the ship’s bow, stern, bridge, and stack in a time series plot
using MATLAB; the background results from TAWS were compared to the
corresponding hourly in-situ radiometric temperature of SST in a separate time series plot
using MATLAB. Note the in-situ measurements were not averaged since this study
wanted to compare the real-time radiometric temperature verses the TAWS predicted
radiometric temperature for the target and the background.
Hourly percentage errors for the target and background were then calculated at
each location for each sensor heading with the assumption that the in-situ radiometric
temperatures were the “actual” or “accepted” value. These hourly percentage errors were
then averaged over the entire time period at each location for each sensor heading and
then compared to one another to determine the inaccuracy of the TAWS predicted values.
Results
At the first location, for a sensor heading of 060° (bow facing), TAWS over
predicted the bow and the stern temperatures and under predicted the bridge temperatures
by less than 1% error; TAWS also under predicted the stack and SST by 2% and 7% error,
respectively (Appendix C). When the sensor heading was 240° (stern facing), TAWS
under predicted the bow, stern, and bridge temperatures by less than 1% error and under
predicted the stack and SST by 3% and 7%, respectively (Appendix D). Note in both
figures of the in-situ target and predicted temperatures at this location that the ship’s
7
stack radiometric temperatures were not close to the TAWS predicted target values from
1600 UTC, February 1, 2005, to 0700 UTC, February 2, 2005. Also, note in both figures
of the SST and the predicted background temperature at this location, TAWS predicted
the background radiometric temperature to be less than the in-situ temperature and below
freezing throughout the entire model period.
At the second location, for a sensor heading of 060° (bow facing), TAWS over
predicted the bow and the stern temperatures and under predicted the bridge temperatures
by less than 1% error; TAWS also under predicted the stack and SST by 3% and 7% error,
respectively (Appendix E). When the sensor heading was 240° (stern facing), TAWS
under predicted the bow, stern, and bridge temperatures by less than or approximately 1%
error and under predicted the stack and SST by 4% and 7%, respectively (Appendix F).
Similar to the first location, both figures of the in-situ target and predicted temperatures
at this location show that the ship’s stack radiometric temperatures were not close to the
TAWS predicted target values for the entire model period. Also, like the first location,
both figures of the SST and the predicted background temperature show that TAWS
predicted the background radiometric temperature to be less than the in-situ temperature
and below freezing throughout the entire model period.
At the third location, for a sensor heading of 060° (stern facing), TAWS over
predicted the bow and the stern temperatures and under predicted the bridge temperature
by less than 1% error; TAWS also under predicted the stack and SST by 3% and 9% error,
respectively, (Appendix G). When the sensor heading was 240° (bow facing), TAWS
under predicted the bow, stern, and bridge temperatures by less than 1% error and under
predicted the stack and SST by 3% and 9%, respectively (Appendix H). Similar to the
8
first and second location, both figures of the in-situ target and predicted temperatures at
this location show that the ship’s stack radiometric temperatures were not close to the
TAWS predicted target values for the entire model period. Also, like the first and second
location, both figures of the SST and the predicted background temperature show that
TAWS predicted the background radiometric temperature to be less than the in-situ
temperature and below freezing throughout the entire model period.
At the final location, for a sensor heading of 060° (bow facing), TAWS under
predicted the bow, stern, and bridge temperatures by less than or approximately 1% error
and under predicted the stack and SST by 3% and 6% error, respectively (Appendix I).
When the sensor heading was 240° (stern facing), TAWS under predicted the bow, stern,
and bridge temperatures by less than or approximately 1% error and under predicted the
stack and SST by 4% and 6%, respectively (Appendix J). Similar to the other locations,
both figures of the in-situ target and predicted temperatures at this location show that the
ship’s stack radiometric temperatures were not close to the TAWS predicted target values
for the entire model period. Also, similar to the other locations, both figures of the SST
and the predicted background temperature show that TAWS predicted the background
radiometric temperature to be less than the in-situ temperature and below freezing
throughout the entire model period.
Discussion
Imaging IR sensors actually respond to the difference in radiance between a target
and its background across a wavelength band that is approximately 8-12 microns for
long-wave infrared (LWIR) sensors and 3-5 microns for mid-wave infrared (MWIR)
sensors. However, it is customary to specify sensor performance in terms of a
9
temperature difference. This practice is made possible by converting the physical (actual)
temperature of the target and its background to individual equivalent blackbody
temperature (EBBT). This conversion specifically involves the IR inband emissivity of
the body.3
Target Comparison
For all locations, TAWS does well at predicting the target radiometric
temperature when comparing them to the measured bow, stern, and bridge radiometric
temperatures. However, the predicted target radiometric temperatures are lower than the
measured stack radiometric temperature for different periods of the diurnal cycle.
TAWS version 3 supports two IR target contrast models: the new Multi-Service
Electro-optics Signature (MuSES) model and the older Target Contrast Model #2
(TCM2).4 The fundamental difference is that TCM2 is a simple 1-D solver that allows
conduction only through the thickness of the material while MuSES is a full 3-D solver
that includes lateral heat transfer. Since MuSES is a more sophisticated model than the
TCM2 model, it performs many more calculations and its time requirement for a model
run is significantly longer. TAWS determines which model to run based on the target or
targets selected for analysis. In this study, TAWS selected TCM2 to predict the target
radiometric temperature.
The thermal model, TCM2, used to compute target temperatures treats the target
as a distinctive three-dimensional network of nodes that exchange heat with one another
and with their environment, both exterior and interior.4 Many phenomena interact to
produce the thermal scene containing the target and its background by which TCM2
3
Air Force Weather Agency Training Division, “Air Force Weather Systems Training Workbook for
Target Acquisition Weapons Software (TAWS) 3.2”. Offutt AFB, NE. August 2004.
4
TAWS Version 3 Help Index, “TAWS Physical Models: IR Model/Target Contrast Model”. August 2004.
10
incorporates into the model. These principle phenomena include solar loading, sky
radiation, spectral bands, and mass and heat transfer.
Solar load is distributed by its components: direct, diffuse, and reflected. This
allows the proper directionality, shadowing, and optical properties to be taken into
account. The sky radiation model inside TCM2 also incorporates directionality,
shadowing, and optical properties. In addition, two spectral bands are treated. The total
band represents the radiative contribution to heating. The LWIR and MWIR sensor bands
treat reflected components in radiometric target and background temperatures; the MWIR
sensor band also includes sky and solar reflections. Finally, mass and heat transfer
effects of evaporation, condensation, sublimation, and precipitation are directly
accommodated into TCM2.5
In TCM2, all emissions and multiple reflections interact correctly in both the total
band and the sensor band through convective coupling of the target to the ambient air.
This process is based on the vector sum of wind speed and direction and the target speed
and heading. Also, in TCM2, a thermal analyzer provides conduction paths as well as
internal heat sources. Fluid-flow heating and cooling effects are incorporated in a
straightforward mass-flow nodal/conductor network. Provisions are made for
conductances, capacitances, and heating rates that depend on time or temperature in
which internal heat sources may also be time-dependent.6
Therefore, TCM2 is a “first principles” model. Laboratory values are used for the
physical properties of any particular model (e.g., nodal values of heat capacity, internodal
5
TAWS Version 3 Help Index, “TAWS Physical Models: IR Model/Target Contrast Model”. August 2004.
6
Johnson, K.R., 1991: Technical Reference Guide for TCM2. Georgia Tech Research Institute, Georgia
Institute of Technology, Atlanta, GA.
11
values of head conductance, coefficients of convective exchanges, etc). These
fundamental attributes are detailed in files that are unique to each target.7
A major advantage of a first principles approach is that it makes possible the
creation of a purely “theoretical” thermal model, just from blueprints and other
engineering specifications of a target. Beyond this, the parameters of the model may be
“fine-tuned” if simultaneous radiometric and meteorological measurements are
subsequently available for the target.7 This could explain why the radiometric
temperature of the ship’s stack was not close to the TAWS predicted values. The larger
percentage errors of the TAWS predicted values in comparison to the ship’s stack
radiometric temperatures could be a result of a lack in simultaneous radiometric and
meteorological measurements for the R/V Point Sur.
Background Comparison
In its model output, TAWS provides the radiometric temperature of the
background for the waveband of the sensor used. It also includes the reflected sky
radiation in the background calculation, which can therefore exhibit diurnal variations.
Therefore, since TAWS assumes that the ocean is an effective blackbody, the SST as
measured by the boom probe (thermometric temperature) should be almost exactly equal
to the temperature measured by a radiometer (radiometric temperature).
Prior to each model run, the meteorological input into TAWS required that the
background thermometric temperature be entered. However, after each model run, the
predicted background radiometric temperatures were not equal to the background
thermometric temperatures. In all cases, the radiometric temperature was substantially
lower than the thermometric temperature, between 10 and 30K less. In addition, when
7
TAWS Version 3 Help Index, “TAWS Physical Models: IR Model/Target Contrast Model”. August 2004.
12
these predicted radiometric background temperatures were compared to the in-situ
background radiometric temperatures, they were lower in all cases.
Several factors can lead to differences between a blackbody’s thermometric
temperature and its radiometric temperature (also known as “effective temperature”).
These factors can affect both absolute and differential temperatures.8
If the output of a blackbody is less than its ideal value, meaning its emissivity is
less than 100%, it will appear to the radiometer to be at a lower temperature. For
example, a blackbody with emissivity of 96% will emit 96% of the flux that an ideal
blackbody would emit. Additionally, it will reflect 4% of the flux incident on the
blackbody surface from the environment. This can be compensated for by increasing the
blackbody’s setpoint slightly, so that the net output is correct.8 Determining just how
much to boost the setpoint is where things get more complex; the correction to the
blackbody setpoint will depend on how much energy the environment supplies.8
Usually, it’s practical to treat the environment as a blackbody at room
temperature, referred to as “ambient temperature”.8 Therefore, if the blackbody
temperature and the ambient temperature are equal, then no correction is needed,
regardless of emissivity, and the radiometric temperature is equal to the thermometric
temperature. But, if the blackbody temperature is more than the ambient temperature, the
radiometric temperature will be less than the blackbody temperature and a correction is
required.8
Some commercial blackbodies attempt to compensate for radiometric losses by
inserting a gain or offset correction into the temperature setpoint. But, the ambient
8
Santa Barbara Infrared, Inc., “Radiometric Temperature: Concepts and Solutions”.
http://www.sbir.com/pdf/RdeltaT.pdf, date not given.
13
temperature dependence and wavelength dependence makes this correction an uncertain
approximation at best.9 Certain corporations such as Santa Barbara Infrared, Inc. (SBIR)
that design and test civilian and military IR systems have blackbodies available with a
more sophisticated radiometric correction option, which measures ambient temperature
and automatically computes and applies a true wavelength-compensated adjustment. As
an alternative to such a blackbody, a user can, using spectral data, ambient temperature
and Planck’s Law, compute a correction and add it to the blackbody’s setpoint, updating
that setpoint as ambient temperature changes.9
Radiometric attenuation is also a source of differential temperature errors. Since
the two temperatures that compromise the difference in temperature (∆T) are each at a
separate distance from ambient temperature, each will require its own correction. Quite
frequently one of the temperatures is “floating” at ambient and needs no correction.
However, if a differential blackbody’s target is heated above ambient by its proximity to
the blackbody, then its temperature will need correction.9
The cause of this error, which affects only differential temperatures, is that
blackbody radiance is not a linear function of temperature. So, a given temperature
difference will not always yield the same radiance difference (Appendix K). A
differential blackbody typically allows one temperature (the target) to “float” with
ambient, and controls the temperature difference referred to this ambient target.9
Therefore, you can see that at a 298K ambient, a ∆T of 5°C will generate a radiance
contrast of 994 µW/cm2, but at 313K, a ∆T of only 4.3°C is needed to generate the same
9
Santa Barbara Infrared, Inc., “Radiometric Temperature: Concepts and Solutions”.
http://www.sbir.com/pdf/RdeltaT.pdf, date not given.
14
contrast. So, if the blackbody is set to 5°C ∆T, it will yield a higher radiance contrast,
and thus a higher apparent ∆T, if the ambient temperature rises.10
This creates a problem for IR systems because the imaging system responds to
radiance contrast, while a blackbody is typically controlling temperature contrast. One
suggestion to accommodate this difference is to consider a quantity called Radiometric
Temperature Difference (R∆T) which is defined as “the same radiance contrast,
integrated over the waveband of interest, which would be generated by a blackbody at
298K and a second blackbody at a temperature of R∆T above 298K”.10 Although R∆T is
expressed in units of temperature, it actually defines a radiance contrast, a phenomenon
that is wavelength-dependent.
Altogether, this phenomenon is not due to emissivity being less than 100% or to
some other limitation of the blackbody, but is simply a consequence of Planck’s Law.10
In order to get an accurate radiance contrast, both emissivity correction and linearity
correction must be performed. This further reduces the utility of the simplistic gain-and-
offset correction for radiometric temperature. The user must either compute the
correction off-line, or use a blackbody controller such as SBIR’s that computes and
applies true radiometric corrections.10
Since TAWS predicted the radiometric background temperatures to be below
freezing and roughly 10% less than the in-situ radiometric background temperatures, it is
apparent that the blackbody controller or lack there of for the ocean in TAWS has major
radiometric correction issues. This can give rise to many potential problems if this
10
Santa Barbara Infrared, Inc., “Radiometric Temperature: Concepts and Solutions”.
http://www.sbir.com/pdf/RdeltaT.pdf, date not given.
15
software is to be used by the U.S. Navy, specifically in target engagements over the
ocean.
CONCLUSION
TAWS can be a very effective tactical decision aid for military EO systems.
However, based on the performance predictions in this study, TAWS requires some
adjustments when conducting such predictions over the ocean. With the proliferation of
this software throughout DoD, modifications are vastly needed in the Target Contrast
Model to account for the radiometric differences between a target and its background that
U.S. Navy assets may encounter over the ocean. Otherwise, TAWS may not be
considered as a credible tactical decision aid among most warfare communities in the U.S.
Navy.
As a result of this study, I recommend that a method to correct these errors be
investigated and found prior to the unlimited distribution amongst the Armed Forces
under the DoD. TAWS should not predict a background to be below freezing unless it
truly is below freezing. Therefore, it is essential that these improvements in the software
happen soon so that the U.S. Navy can reap the benefits of such an advanced EOTDA.
ACKNOWLEDGEMENTS
Special thanks to Dick Lind, NPS meteorologist and technician, who graciously
configured the IR probes and the data recorder on the R/V Point Sur prior to the student
cruise and who also gladly answered the numerous technical questions I had throughout
this study.
Special thanks to Jerome Hernandez, Capt, USAF, who gave me the idea for this
project and who also graciously advised and assisted me throughout the entire study. I
16
cannot express my personal gratitude enough to him for all the time he devoted and for
the patience and guidance he demonstrated for what initially seemed like a simple project.
Additional thanks to Professor Ken Davidson, Department of Meteorology, Naval
Postgraduate School, for his guidance and support in my efforts to test TAWS
capabilities over the ocean.
REFERENCES
Air Force Weather Agency Training Division, “Air Force Weather Systems Training
Workbook for Target Acquisition Weapons Software (TAWS) 3.2”. Offutt AFB,
NE. August 2004.
Johnson, K.R., 1991: Technical Reference Guide for TCM2. Georgia Tech Research
Institute, Georgia Institute of Technology, Atlanta, GA.
Santa Barbara Infrared, Inc., “Radiometric Temperature: Concepts and Solutions”.
http://www.sbir.com/pdf/RdeltaT.pdf , date not given.
TAWS Version 3 Help Index, “TAWS Physical Models: IR Model/Target Contrast
Model”. August 2004.
17
THIS PAGE INTENTIONALLY LEFT BLANK
18
APPENDIX A
TAWS Model Process for IR Performance Predictions
User Model Visual Representation of Component Products
Target Illumination
Information Model
(Intelligence) Range = 0
Target Scene
Contrast
Background Temp = 300
Model
Meteorological Target Temp = 310
∆T=10
Information
(Weather Team) Range = 1 2 3 4 5 6 7 8 9 10 11
IR
Transmission
Model
Sortie ∆T=10 ∆T=8 ∆T=6 ∆T=4 ∆T=2 ∆T=1 ∆T=0
Information
(Aircrew)
Sensor WFOV TAWS Prediction
Performance MDT Difference = 4
Model
WFOV MDT
NFOV Detection Range = 5.5
MDT Difference = 1
NFOV MDT
Detection Range = 9
TAWS
Prediction
Shows schematic diagram of the model process in TAWS for IR performance predictions. The
focus of this study specifically concentrates the “Target Scene Contrast Model”, one major portion
in the overall modeling process. Note that this diagram was designed by Jerome Hernandez, Capt,
USAF.
19
APPENDIX B
Target Heading: 060°
TOT: 1100, 3 Feb 05
Target Heading: 240°
TOT: 0800, 2 Feb 05
Target Heading: 240°
TOT: 2200, 1 Feb 05
TOT: 2200, 5 Feb 05
Shows the TAWS-generated target tracks used for this study at four different times during the OC3570
cruise. Note that the first and last tracks are shown above as in the same location. Although, these two
tracks were relatively close to one another, they were not exact.
20
APPENDIX C
Shows the target and background comparisons at the first location, sensor heading 060° (bow
facing), from 1000 UTC, February 1, 2005, to 0700 UTC, February 2, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
21
APPENDIX D
Shows the target and background comparisons at the first location, sensor heading 240° (stern
facing), from 1000 UTC, February 1, 2005, to 0700 UTC, February 2, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
22
APPENDIX E
Shows the target and background comparisons at the second location, sensor heading 060° (bow
facing), from 1000 UTC, February 2, 2005, to 0500 UTC, February 3, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
23
APPENDIX F
Shows the target and background comparisons at the second location, sensor heading 240° (stern
facing), from 1000 UTC, February 2, 2005, to 0500 UTC, February 3, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
24
APPENDIX G
Shows the target and background comparisons at the third location, sensor heading 060° (stern
facing), from 1000 UTC, February 3, 2005, to 0100 UTC, February 4, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
25
APPENDIX H
Shows the target and background comparisons at the third location, sensor heading 240° (bow
facing), from 1000 UTC, February 3, 2005, to 0100 UTC, February 4, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
26
APPENDIX I
Shows the target and background comparisons at the last location, sensor heading 060° (bow
facing), from 1000 UTC, February 5, 2005, to 0700 UTC, February 6, 2005. Note in the target
plot ‘WFOV’ and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the
background plot ‘Background’ is the TAWS predicted radiometric temperature.
27
APPENDIX J
Shows the target and background comparisons at the last location, sensor heading 240° (stern facing),
from 1000 UTC, February 5, 2005, to 0700 UTC, February 6, 2005. Note in the target plot ‘WFOV’
and ‘NFOV’ are the TAWS predicted radiometric temperatures and in the background plot
‘Background’ is the TAWS predicted radiometric temperature.
28
APPENDIX K
Shows the in-band blackbody radiance as a function of temperature. Notice that blackbody
radiance is not a linear function of temperature; a higher ambient temperature requires a smaller
∆T to generate the same radiance contrast.11
11
Santa Barbara Infrared, Inc., “Radiometric Temperature: Concepts and Solutions”.
http://www.sbir.com/pdf/RdeltaT.pdf, date not given.
29
Related docs
