`
`Distribution authorized to US. Government
`agencies and their Contractor's, Critical Technology.
`13, March, 199$ Other requests for this
`document shall be referred to ONR
`800 North Quincy St., Arlington VA 22217-5666
`
`Hypervspectral Imaging and Infrared Spectroscopy Using Pacific Advanced Technology’s
`Image Multi~spectral Sensor (IMSS) and Amber Engineering’s Radiance 1 Camera.
`
`Michele Hinnrichs, Mark Massie
`Pacific Advanced Technology
`1623 Mission Dri, Suite 3
`Solvang, CA 93464-0679
`
`.ieff Frank
`
`Amber Engineering
`A Raytiteon Company
`5756 Thornwood Dr.
`
`Goleta, CA 93111
`
`ABSTRACT
`
`IMSS utilizes a very simple Optical design that enables a robust and low cost hyper-spectral imaging
`instrument. This technology was developed under a phase II SBIR with the Air Force Philips Lab. (Dru
`Paul LeVan) for the midwave infrared to perform clutter rejection and target identification based upon
`infrared spectral signatures The prototype instrument has been field tested on numerous occasions and
`successfully measured background, aircraft and missile plume spectral
`PAT currently has several
`contracts to commercialize this technology both for the DoD and the commercial market. Under contract
`to the BMDO, (Paul McCarley), with matching funds from Amber Engineering we are developing an
`F/2.3 system that will be sold by Amber Engineering as an accessory to the Radiance 1 camera, PAT is
`also under contract with ONR, ( Mr. Jim Buss) to develop a longwave infrared version of M88 as well
`as a MWlR version tuned to operate as a “Little Sister Sensor for target identification" for the Navy’s
`IRST’s,
`
`The purpose of this paper is to briefly describe the hyper—spectral image data that was collected in the
`field at Long .iump ‘94, and Santa Ynez Peak using the IMSS prototype hyper-spectral imagery. Examples
`of spectral images as well as spectra of different aircraft at various ranges, power settings. and aspect
`angles. an Atlas liquid hydro—carbon burning missile and a solid booster. All data presented in this paper
`are a result of a single spectral scan" The limitation in digital storage of the prototype system do not
`allow multiple scans in order to improve signal to noise In spite of this limitation the performance of the
`prototype system has proven to be excellent.
`
`BACKGROUND
`
`Pacific Advanced Technology (PAT) has deveIOped a prototype laboratory system, lMSS, a hyper—
`spectral
`imaging system What originally began as a Phase I SBIR to the Air Force Space Division,
`Space and Missile Command,
`for detection and identification of ballistic missile in a cluttered
`
`1
`
`UNCLASSIFIED
`
`LEAK SURVEYS, INC
`
`EXHIBIT 2031
`
`FUR V- LSI
`TRIAL |PR2014-00411
`
`
`
`UNCLASSIFIED
`
`background was successfully carried to a Phase ll SBIR program with the Air Force Phillips Lab.
`Kirtland AFB.
`
`the IRIS Targets,
`imaging have been reported at
`Capabilities of this approach to hypermspectral
`Backgrounds and Discrimination meeting group' as well as the IRIS Passive Sensors working groupz.
`Presentations were also made in August of 1994 at the International Symposium on Spectral Sensing
`Research '94 (ISSSR)3 and the 3rd Annual AIAAIBMDO Interceptors meeting in San Diego“.
`
`to the Office of Naval
`Pacific Advanced Technology is currently under an SEER Phase II contract
`Research, Jim Buss. to develop the next generation of lMSS, This system will use a 5l2 x 512 InSb
`"region of interest" FPA developed by Amber Engineering under contract to BMDO and an f/2.3 lens.
`The smaller pixel size of the 512 x 512 FPA coupled with the {/23 lens will allow finer spectral
`resolution. The “region of interest" readout will allow the measurement of temporal as well as infrared
`Spectral signatures of targets. The intent of this program is to deveIOp a “Little Sister Sensor" for the
`Navy’s IRS? to be used for target identification
`
`Pacific Advanced Technology is also under contract to the BMDO, Paul McCarley. with matching
`funding from Amber Engineering to commercialize the IMSS technology for the DOD and commercial
`market place“ PAT is developing a MWIR hyper—spectral radiometric imager as an accessory to Amber
`Engineering's Radiance 1 camera. The system will match the Radiance l {723 cold shield with a
`nominal focal length of 65 mm. The lMSS optics will be coupled with an afocal system to allow for a
`larger collecting aperture. The predicted spectral resolution of ARM. 2 0.25 % over the full spectral band
`from 3 to 5 tun.
`
`The lMSS lens system for the Radiance 1 camera will have a volume of approximately 150 in3 and will
`weigh approximately 2 lbs. The spectral data collection, storage and processing will be performed using
`Amberview and a WindOWS based spectral data evaluation software developed by PAT.
`
`
`
`IHinrtrichs, M.. Massic. M, " A New and innovative Instrument for Infrared Imaging Spectroscopy" IRlS TED
`meeting in Monterey. CA. Feb. 1994.
`2Massie. Mi. Hinnrichs. MA, "Hypenspectral Imaging Radiometer Using Staring 128 x 1'28 lnSb Focal Plane Array
`and Diffractive Optics." lRlS Passive Sensor meeting in Albuquerque, NM... March l994.
`3 Massie, M. Hinnrichs. NI, “An Infrared Hyper-Imaging Spectrometer for Atmospheric Studies and Environmental
`Remediation", San Diego July l994.
`4Massie, M..
`llinnrichs. M., “An Infrared Hyper-imaging Spectrometer for Missile Seekers", AlAA/BMDO
`Interceptor Technology Conference. San Diego, July l994.
`
`2
`
`UNCLASSIFIED
`
`
`
`UNCLASSIFIED
`
`DESCRIPTION OF THE IMSS INSTRUMENT
`
`imaging spectrometer using a new and
`is a dispersive type hyper—Spectral
`The lMSS instrument
`innovative technique invented by Pacific Advanced Technology, “Image Multi—spectral Sensing", The
`IMSS prototype instrument uses a 128 x 128 InSb FPA built by Amber Engineering as the detector. The
`prototype configuration of the IMSS instrument is tuned for a peak performance at 3,9 ttm with a 17/35,
`160 mm focal length lens system,
`spectral coverage from 3.0 to 50 um, and a measured Spectral
`resolution 0(1) AM}. of 0.41%, (6 crud), The IMSS can step through approximately 320 spectral bins per
`micron The field of view of the instrument is 25 degrees with and IFOV of 0.3 mrad. The instrument
`can collect a full spatial spectral cube of 128 x 123 pixels from 3 to 5 mm in less than 2 seconds, The
`noise equivalent Spectral irradiance, (NESR), has been measured at 5.52 x 10'? w/cmzusr—ttmi
`
`A picture of the IMSS prototype instrument is shown in figure I,
`
`
`
`Figure 1. The prototype Image Mani-spectral Sensor
`
`The IMSS approach to image hprFSPBCIFOSCOpy utilizes a very simple diffractive optical system as
`contrasted with the complexity of the more traditional approach.
`It is also lightvveight, low cost and
`relatively insensitive to normal vibration effects inherent in an airborne platform,
`
`EXAMPLES OF PERFORMANCE
`
`The IMSS instrument has gone through extensive characterization in the laboratory at PAT prior to field
`testing, An example of the performance is represented in figure 2, where the Spectrum of a butane flame
`for each of two fuel—air mixture ratio conditions is shown, Each of the plots in the figure represent the
`average of three spectral runs at the indicated condition, The strongest spectral features are the C02
`emission lines, the blue and red spike, showing in the 4.2 and 45 pm region, and the unique butane
`
`3
`
`UNCLASSIFIED
`
`
`
`UNCLASSIFIED
`
`hydrocarbon spectrum at 3.4 ttm. This data was taken using 320 spectra! bins per um, 1/2 of the fut!
`spectra! resoiotion of the IMSS system. The difference between a clean and a dirty burning butane flame
`is clearly seen in the spectral region between 3.0 to 3.7 ttrn as shown in figure.
`
`Dirty & Clean Burning Butane Flame
`
`4500
`
`point Source
`
`320 Bins/Micron Spectrai Resolution
`
`Frame rate 217
`integration Time 4.03 ms
`
`4000 *
`3500
`g
`g 3000
`E 2500
`$
`g
`2000
`a...
`E 1500
`E000
`
`_ _
`
`500
`
`0
`
`2.9
`
`“L
`
`3.1
`
`V .
`
`i
`
`.
`
`€02
`
`t
`
`--——m- Dirty
`V ”bClean
`
`Hints th.
`l
`
`PF
`
`I"
`
`M
`
`’4-
`
`3.3
`
`3.5
`
`3.7
`
`3.9
`
`4.1
`
`4.3
`
`4.5
`
`4.7
`
`Wavelength (Microns)
`
`Figure 2. Butane flame spectra for two fuel-air mixture ratio conditions as measured using the IMSS
`instrument.
`Two Spectral images of a dirty burning butane flame are shown in figure .3. The first image is for :1
`Spectral bin centered at 3.40 pm and the second is at 4.57 tun. Notice the difference in shape for the two
`spectral regions. The two distinct spectral
`lobes in figure 3 a) are characteristic of all dirty burning
`butane flames in this spectral region.
`
`
`
`a)
`
`b)
`
`Figure 3. Spectral images of a Butane flame, a) for a spectral bin centered at 3.40 mm, b) a spectral bin
`centered at 4.57 mm.
`
`4
`
`UNCLASSIFIED
`
`
`
`UNCLASSIFiED
`
`LONG JUMP ‘94
`
`FIELD MEASUREMENTS
`
`The PAT lMSS hypernspectral imaging instrument participated in Long Jump 94'. Long Jump is held at
`the Barcroft Laboratory test site, located at an elevation of 12,500 feet in the White Mountains near
`Bishop, California. The location allows the sensors to be placed at nearly the same altitude as the test
`aircraft as the aircraft approach and exit the site. This location also allows the aircraft to descend below
`the earth’s horizon into the background clutter. The N38 hyper—spectral camera was mounted on a
`Contraves tracking mount and measured the Spectral signature of numerous military and commercial
`aircraft both fixed wing and helicopters. The simplicity and robust nature of M38 coupled with the
`ability to perform tracking of the targets lead to an extremely successful data collection exercise To our
`knowledge PAT was able to collect the most complete imaging spectral data set on various types of
`military aircraft that exists today The imaging spectral data set consists of about 1‘5 GBytes of data,
`The testing was funded by a Phase Ii SBIR program with the Air Force Philips Lab.
`
`The data that was collected at Long Jump 94' using the IMSS hyper—spectral imager consists of:
`0 Aircraft type
`Range
`Altitude
`
`9aOC
`
`Power setting
`Aspect with respect to sensor
`Weather data
`
`0
`9
`
`Spectral signature from 3 to 5 ttm
`lRIG timing data"
`
`Long Jump ‘94 data collected consisted of numerous aircraft There are 12 different aircraft and in each
`data set there are spectra for different aspects; nose. tail and side. There are spectra of both unresolved
`and resolved images“ The maximum range is out to 25 mini" Figure 3 shows two narrow spectral band
`images, (031 ttm), of a resolved tactical aircraft with a nose aSpect. (spectral band centered at 4548 um).
`and close range plume, (spectral band centered at 3102 um).
`
`Figure 4 shows an example of an a tactical aircraft at 23 nmi, tail aspect in afterburning power setting
`This also is an example of the range performance of IMSS. It is interesting to compare the spectrum of
`this same aircraft in afterburner with the same type aircraft but different run in MIL power shown in
`figure 5" The ratio of the red and blue spike to the continuum caused by the soot is very different. Also
`there is a strong line at 3J9 ttrn.
`
`5
`
`UNCLASSIFIED
`
`
`
`UNCLASSIFIED
`
`
`
`Figure 3. Narrow bond spectral images of tactical aircraft at close range, a) at 4.458 um shows the inlet to
`the turbines and b) at 3.102 um shows the plume, (two center bright regions) plus the exhaust from the
`environmental control system, (the two outside bright regions).
`
`Tactical Aircraft at 23 nml AB Spectrum
`
`3.1945
`
`4.1749
`
`1665+”
`
`1.64E+14
`
`1.62E+14
`
`1n60E+14
`
`1n58E+14
`
`1.‘56E+14
`
`1.54E+14
`
`L52E+14
`
`1.50E+14
`
`
`
`Signal@Detector(phlemZ-s)
`
`3
`
`3.2
`
`304
`
`3.6
`
`4
`3.3
`Wavelength (microns)
`
`4‘2
`
`4.4
`
`4.6
`
`4.8
`
`Figure 4. An example spectrum ofa tactical aircraft at 23 nmi tail aspect in afterbumer.
`
`6
`
`UNCLASSIFIED
`
`
`
`
`
`.53banana:Eu"...22m
`
`mammaw‘mog
`
`
`
`:58:“Sauna.
`
`3.5.-M:55
`3%”_Ea
`
`ufiafima2:»Eu...-3”imam."
`
`
`
`I3+ng.a«56$
`
`
`
`«383.3gauze—urn;
`
`«H.535;3.wein5.a.maWminvanna33..3.vm.nWm3..finnu.n.n..V«153mm
`
`
`m.:53.”
`
`Ann—nuns:
`
`..3&qum
`
`4E+mn3a
`
`_m3&8."
`
`
`
`
`
`
`
`:53;.535.:EE?325:.
`
`
`
`md3.3a...u3.Wmtn3..n3.ad5“3.uana...”“an3”m_...3&3."mV.....2&2."
`
`
`
`fl
`
`.55£8;a:5fine;
`bran:*wtaxa:a«3.5:;|3&chItuna-HE.Iwing."
`
`
`
`£45:_m34“.:A35.3.1«$35m3:3:3&8.“
`3+mofiflM4.vfi+wafimmI3+3?um.Sing
`
`
`
`
`
`
`
`“H.333”.Sankara},mun—rut“:Euaguuui.
`
`we3.X.«a.wanma1m2..nqwdad3a.»vm4.”3”«anadn...+uL.
`
`
`
`3:53”fiiupxkaccuflEvEwuumgag
`3m»:m3&3.“
`
`+.e
`
`l+..w
`
`and:..m.
`
`
`
` IE+m¢3IEnaiu«Hamcflamffiuufl.(.E2.:I352.1"
`753.:mL.+.
`
`
`htauz.3qu_m.“WNW.”_2mm:
`
`I«$333.5.2.2
`1—V3”wmmm:0._.VH+MEA
`
`
`W2mm”.m332»H3333
`
`
`
`Eras."H.331
`
`ummwnm
`
`“m3&2..."
`
`E!
`"WNdeI-“IIG a vmfiIs
`
`(I
`inundilnmm e Nazis
`
`{:
`-1mmd)mpapu @ 5311319
`
`
`
`
`
`a.nanny”new«amme23magmaENE33..Eummmfiwmmawe2&5me.mEmma
`
`
`
`
`
`ngm<dUZDh
`
`
`
`
`UNCLASSIFIED
`
`SANTA YNEZ PEAK
`
`imager was taken to Santa Ynez Peak in the California coastal range at an
`The IMSS hyper'~spcctr'al
`altitude of 4.300 feet over looking the Santa Barbara channel about 35 nmi. from Vandenberg AFB”
`While Spending several nights at this location waiting for the launch of a solid booster missile PAT took
`the opportunity to collect hyper—spectral imagery and spectra of targets of opportunity Some of these
`spectra are shown in figure 6.
`
`The top left is the spectrum of a single engine Piper Dakota aircraft at a range of L3 nmi and 4000 foot
`altitude. This spectrum has a gray body shape The spectra in the upper right is from a flame burning on
`an oil rig in the Santa Barbara channel and its reflection at a range of about 13 nmi, The middle left is
`the spectrum from a refinery plant at the base of Santa Ynez Peak at a range of about It) nmii Although
`both signals looked like bright sources in the infrared image they have very different spectra. The one
`that has the characteristic spectrum of a hydrocarbon flame was from a smoke stack (anomaly at 4.0 urn
`is due to an occultation of the target by a chain link fence at that spectral region field of view) and the
`other was most likely a hot gray body source. The Spectrum on the middle right is for the setting sun.
`Notice that the solar signal doesn't contribute significantly above 3.4 ttmn The spectrum on the loWer left
`is a solar glint of a cloud The IMSS instrument blurs the spectral image ol‘extended objects like clouds
`and averages the spectral signal, this is why the typical C02 absorption region is missing in this spectrum.
`This is an advantage of
`IMSS for clutter rejection There is a dramatic difference between the target
`spectrum and that of the cloud as observed with IMSSi On the bottom left is a spectral image of the
`cloud background taken at 4.0 ttmi
`
`On December 30 at 2:01 AM, after eight delays. IMSS was used to measure the spectral signature of an
`Atlas booster: The IMSS sensor was located in the Santa Ynez Valley at an elevation of about 800 feet,
`The range to the launch pad was about 32 nrni. and the missile initially came closer to the sensor as it
`was launched into orbit. The Atlas has a liquid burning hydro—carbon engine and burns fuels similar to
`that of tactical aircraft, Several spectral
`images for the Atlas are shown in figure 7. The measured
`spectrum of the Atlas is shown in figure 8. The first spectrum is just after lift off when the missile is
`only a few hundred feet off the ground The second two spectra were taken 1 minute later when the
`missile was at an altitude of several thousand feet A spectrum from the head of the plume as well as a
`spectrum from the tail of the plume are shown Notice the significant decrease in the noisy signature in
`the range from .3 to 4.2 um in the tail section of the plume.
`It is believed that this is due to the fact that
`there is less soot in this region.
`
`The IMSS data is compared with data taken by Arnold Engineering Development Center from Santa
`Ynez Peak on September 17, 1986 with an 13’1”th The AEDC spectrometer was located 39 nrni. from
`the launch pad and the IMSS was located about 32 nmi, from the launch pad The comparison is shown
`in figure 9‘ The IMSS spectrum was taken with a single scan. There was no attempt to improve the
`signal to noise performance by taking multiple scans. The reason is that the lMSS prototype system is
`limited in digital data storage and can only store 400 bins of data per data collection cycle The
`similarity between the two spectral is an indication of the remarkable performance of IMSS considering
`its simplicity in design“
`
`
`5 AEDC «TR-8744 supplied to Pacific Advanced Technology by Bill leffrey with IDA.
`
`8
`
`UNCLASSIFIED
`
`
`
`
`“3.»MWNHat.598“Vu+m~¢mum(
`aim?m"can3.gigs._.m,
`
`
`
`
`
`
`
`
`
`«Eaflaw—“£3.15
`
`WVvWmmjEfififimum-
`
`§@_E+M¢HH
`an
`
`
`
`and...“Many.89mm596mmqfifit.»EMSSE.3.53%.«c335mewwhammm
`
`
`
`
`
`mgwm¢flugm
`
`maawmfidug
`
`ER9nun—um
`
`mQZDOmUMufim“wmhmwfik
`
`‘Haims."@x._..1..c\1.:....,y,m.
`
`
`
`m6.vmummm6.vWmS.n..vm+mcc.mPM.W
`
`Egan—sunk
`
`2:3fimcuwufls
`
`an.mM3&8;mWm.
`
`
`
`
`
`”Maugham!IIv~+mmwéW:E.:1?mamOiunEad5:333»F”wigs.”m.E235;
`
`
`
`
`
`
`
`nafitamumfifim(“A34«amtmf
`
` I.m_2&3hm“35ml...2&8.”m
`
`m6vWmmvWmj._9&3“..w.
`
`
`uunnas““3555_m“:3«am.3
`.BQNZWEoEUml«.133m.«35ml!E:S53%
`
`Muggfimm,ugh:w‘b90:55.3
`
`i-ww4—-—I— ‘-
`
`m
`
`3.5254
`
`mn+mcwd
`
`mH+mE=d
`
`mimocfi
`
`mmmgw
`
`(s-fluamd) Janna“ g) “"1315
`
`
`
`vac—U1......A.W53nun—5.8%Eh.m."”use.5320s.VH+mmmMIHw..25.:59.305.»
`
`
`
`
`
`UNCLASSIFIED 3.3671
`3.0132 4.5691
`
`3.1760
`
`4.1911
`
`3.6959
`
`
`
`
`
`Figure7.Typicalspectra]imagesoftheAfiasboosterplumeasitwaslaunchedcutofVandenbergAFBonDecember30.1994.
`
`
`
`
`
`
`
`
`
`10UNCLASSIFIED
`
`
`
`UNCLASSIFIED
`
`um!-
`tun"
`
`tum:
`
`PLUM E A1” ALTITUDE
`vmnmnunc an!
`eat: :11. am
`
`than Him-1|
`lmutlnlrgrllien
`
`mun»! mi
`[Minn-1,05“
`
`
`
`sum
`
`WMmI-Iu—l
`1nuuununnuguuuuuuu
`
`!
`
`I
`I
`‘7
`II
`£8
`f".
`I
`"‘ "‘ "“
`“u” u ”‘1'
`
`f.
`
`I
`'“‘"“'“”
`
`Figure 8. Atlas launched out of Vandenberg AFB on Dec. .30, I994.
`
`AT‘l-Lfi
`
`AT ALTTTU’Dt
`
`¥ mm
`m0"
`
`LII-II
`
`LII‘M
`
`”a."WWW
`nuuuuuuun¢uuuuuqt
`n-fl
`
`'
`
`Figure 9. Comparison OHMSS spectral data for an Atlas iaunch from Vandenbcrg on December 38, 1994
`with AEDC FTIR spectral data taken from Santa Ynez Peak On September 17, 1986"
`
`As a comparison of different types of rocket engine fuels the IMSS instrument measured the spectrum of
`a solid booster. The launch pad was at a range of greater then 50 nmi from IMSS and the missile was
`sent dowa range thus traveled away from the senses The IMSS instrument had an integration time of
`about 2 msec and collected 400 spectral images from 3.0 ~ 4“? pm. The signal at the detector in ph/ch—s
`as a function of wavelength is
`shown in figure it). Notice the difference in this spectrum to that of
`hydrocarbon burning engines" The absence of the noisy signature, (soot), between 3 and 4 tint
`
`I 1
`
`UNCLASSIFIED
`
`m."
`lllhll
`
`Pummsmrr
`eschew
`“immune are
`
`won-arm
`
`lit-36
`this"
`
`lull-cu
`
`Luau
`
`Layla
`
`“IMHO-M rlI
`”m" lwl‘
`
`
`
`UNCLASSIFIED
`
`Laundml It
`Vlndcnlaur: AFB
`1995
`
`use-n
`
`I’ll-H
`
`Lin-ll
`
`J Hum
`o
`115:!“
`
`Run: > 9i] and
`laugh-tine 1mm:
`
`am wwmmw
`use.”
`ILilli-ELISJSJSJJJIJILII—luuuljdj
`\Vlnlmrth {Mk-mm]
`
`Figure 10. Solid booster plume signature at a range greater the 50 nmi. as measured using IMSS.
`
`FUTURE PLANS
`
`The IMSS technology does not require sophisticated and powerful signal processing to pull a spectrum
`out of the data like FI'ER spectrometers do. Therefore. due to the simplicity of the signal processing
`required it is possible to perform signal processing in the analog domain The ultimate hyper—spectral
`sensor basod on the IMSS technology will perform all spectral extraction and signal processing functions
`on the focal plane in real time. The staff a PAT based upon past experience with Smart Infrared Focal
`Plane technology will use these tools to implement
`real
`time Spectral extraction for IMSS. Smart
`Neuromorphic Infrared Focal Plane (SNIF) technology has been previously demonstrated by the staff at
`PAT".
`lMSS technology has been proven to be a simple low cost approach to hypenSpectral imaging.
`The ultimate simplicity of the optics leads toward the implementation of a miniature hyper-spectral
`imaging and spectral analysis system. The combined technologies with the use of on focal plane analog
`signal processing and data buffering operations will lead to the development of an instrument which will
`produce infrared imagery as well as infrared spectra of objects in the field of view.
`
`Currently under contract to ARPA. (Ray Balcerak), a spectrahprocessing focal plane array is being
`designed at PAT which computes the iii-band spectrum of a region of interest (ROI) during the off—chip
`multiplexing operation. At the end of the ROI readout operation, the detected peak value is read off—chip
`from the sample~andwhold capacitor. This peak value represents the peak spectral intensity in the ROI at
`the waveband represented by the current IMSS frame.
`
`The chip can also be operated as a conventional infrared imager, by throwing a switch the readout
`processing integrated circuit (ROPIC) will function as a standard 128 x 128 staring focal plane array. A
`user programmable digital word contains the center pixel location and the window dimension of the ROI
`within the 128 x 128 array. The ROI dimension can range from a .3 x 3 up to a .32 x 32 pixel window.
`With a reduced ROI window dimension, the readout frame can be increased. due to the reduced number
`
`
`.lohnson, R. F. Cannata, W. J. Parrish,
`.I. T. Woolaway, B. L. Huynh, G. A.
`GMassie, M. A.
`"Neuromorphic Infrared Focal Plane Pederms On~Plane Local Contrast Enhancement, Spatial and
`Temporal Filtering", Proc. of IRIS, Passive Sensors, January, 1992.
`
`l2
`
`UNCLASSIFIED
`
`
`
`UNCLASSIFIED
`
`of signals required for multiplexing, At the maximum ROI dimension of 32 x 32, a readout frame rate of
`over 380 Hz is achievable by multiplexing the signal through two separate outputs This frame rate
`approximation assumes a conservative master clock frequency of 500 KHz and over 200 clock cycles
`overhead processing time in addition to the readout time In addition, the 128 x 328 detector sample—and—
`hold buffer allows simultaneous next-frame integration during readout operations.
`
`CONCLUSION
`
`The hyper-spectral camera developed under contract to the Air Force Phillips Lab and field tested with
`funding from both the Air Force Phillips Lab and the Office of Naval Research has shown that this
`instrument is robust, rugged and can measure target and background spectra with relative ease The
`instrument originally developed as a proof of concept laboratory system has been taken into the field on
`several occasions. Once to an altitude of 12,500 feet in the White Mountains of California, then 4,300
`feet at Santa Ynez Peak in the coastal range of California, to various locations in the floor of the Santa
`Ynez valley and to the Lockheed Santa Cruz test facility for field measurements. At Long Jump '94 it
`collected multiple spectral
`images of every flight for over a week At Santa Cruz it collected a
`tremendous amount of diurnal spectral images of targets and backgrounds for three days It was taken
`into the field to collect missile launches out of Vandenberg AFB, making measurements in below
`freezing weather never once had a malfunction All spectra ware taken with a single scan from 3 to 477
`um, due to limitations in digital storage of this prototype system multiple scans were not performed to
`improve the signal to noise“ This is not a limitation of the technology and more digital storage will be
`incorporated in future systems.
`
`Under contract to ARPA we are now designing an all analog on focal plane processing circuit which will
`allow the IMSS to perform hyper-spectral imaging in real time in a light weight low cost package” We
`are commercializing the system with the Radiance 1 camera. Because of the robust light weight nature of
`the IMSS hypenspectral imager it can easily be designed to fly in an RPV as well as in any airborne
`platform for multiple applications such as collection of signatures on targets and backgrounds, for threat
`warning, surveillance, clutter rejection. target identification. pollution monitoring and drug interdiction.
`
`i3
`
`UNCLASSIFIED
`
`