Through a US Army sponsored SBIR project we are developing terahertz components based on integrated GaAs Schottky diodes for the frequency range from 200 - 700 GHz. These new components are inherently broadband and therefore require no mechanical tuners. Rather, they can be electronically swept across significant frequency bands and are therefore useful for chemical and biological spectroscopy. This talk will focus on our demonstration of a terahertz frequency Transmit / Receive capability which may be of use for CB detection and secure communications.

Significant progress has been achieved during the last several years relating to experimental and theoretical aspects of Terahertz (or Sub-millimeter wave) Fourier transform spectroscopy of biological macromolecules. Multiple resonance due to low frequency vibrational modes within biological macromolecules have been unambiguously demonstrated. However till now only solid films of bio-materials have been used for experimental characterization in this spectral range since it was common opinion that high water absorption will prevent from receiving the information on bio-molecules in a liquid phase. At the same time, all biological function of DNA and proteins take place in water solutions. In this work spectra of DNA samples and proteins have been measured in liquid phase (gel) in a spectral range 10-25 cm^-1 and compared with spectra obtained from solid films. The results demonstrate that there is almost no interference between spectral features of material in test and water background except for the band around 18.6 cm^-1. Much higher level of sensitivity and higher sharpness of vibrational modes in liquid environment in comparison with solid phase is observed with the width of spectral lines 0.3-0.5 cm^-1. Gel samples demonstrate effects of polarization. The ability of THz spectroscopy to characterize samples in liquid phase could be very important since it permits to look at DNA interactions, protein-protein interactions in real (wet) samples. One demonstrated example of practical importance is the ability to discriminate between spectral patterns for native and denaturated DNA.

Biopolymers such as proteins, DNA and RNA fold into large, macromolecular chiral structures. As charged macromolecules, they absorb strongly in the terahertz due to large-scale collective vibrational modes; as chiral objects, this absorption should be coupled with significant circular dichroism. Terahertz circular dichroism (TCD) is potentially important as a biospecific sensor, unobscured by spectral features related to abiological material. We have constructed atomistic simulations and elastic continuum models of TCD. These models estimate the magnitude of the TCD and the relation between TCD spectroscopic signatures (zero crossings) and the structure, charge distribution and mechanical properties of biomaterials. A broad band TCD spectrometer based on a polarizing interferometer is developed to explore TCD in biomolecules in aqueous solution. Preliminary results on TCD in lysozyme in water at several terahertz frequencies is presented.

The terahertz regime (0.1 to 10 THz) is rich with emerging possibilities in sensing, imaging and communications, with unique applications to screening for weapons, explosives and biohazards, imaging of concealed objects, water content and skin. Here we present initial surveys to evaluate the possibility of sensing bacterial spores and chemical material using field-deployable electronic terahertz techniques that use short-pulse generation and coherent detection based on nonlinear transmission lines and diode sampling bridges. We also review the barriers and approaches to achieving greater sensing-at-a-distance (stand-off) capabilities for THz sensing systems. We have made several reflection measurements of metallic and non-metallic targets in our laboratory, and have observed high contrast relative to reflection from skin. In particular, we have taken small quantities of materials such as dimethyl methylphosphonate (DMMP) and several variants of Bacillus spores, and measured them in transmission and in reflection using a broadband pulsed electronic THz reflectometer. The pattern of reflection versus frequency gives rise to signatures that indicate specificity of the target. Although more work needs to be done to reduce the effects of standing waves through time gating or attenuators, the possibility of mapping out this contrast for imaging and detection is very attractive.

The potential container penetrating capabilities of THz radiation leads to possible applications for container penetrating sensors for biological hazards. Such an approach requires the presence of distinct THz frequency resonances in the target compounds coupled with sufficiently transparent container materials to allow through container sensing. The results of a THz spectroscopic survey of container and clothing materials are presented along with spectra of materials that were chosen as simulants and markers for illicit biological substances. The spectroscopic data presented show at least partial transparency for materials commonly used for clothing and packaging. We also measure distinct spectral signatures in dipicolinic acid, calcium dipicolinate, peptidoglycan, and 2,6-diaminopimelic acid, biologically significant molecules that are indicative of hazardous spore forming bacteria. These spectra differ significantly from those of the container materials to provide a potential contrast mechanism which could be used for identification.

Gas sensing and identification in far infrared or THz band is useful because many polar molecules have unique spectral fingerprints in this range, which are from the rotational transitions of the molecules. We have investigated the potential of THz time-domain spectroscopy (THz TDS) as a quantitative analysis technique for gas sensing. Ammonia vapor has been chosen as a sample gas. The absorption cross section at 0.572 THz of ammonia in the pressure range of 0.2-20 Torr was extracted to be (5.7±0.3)×10^-20 cm²/molecule. In addition, a pressure calibration curve based on pure ammonia measurements was obtained. Using this calibration curve, we made quantitative analysis on the mixture of ammonia and air at 100 Torr. The result shows that THz TDS is an appropriate technique for quantitative analysis of polar gas and gas mixture. We measured the THz spectra of ammonia at different partial pressures in ~590 Torr nitrogen (78% nitrogen in atmosphere), and obtained a pressure calibration curve. THz spectra of ammonia at different partial pressures in 760 Torr atmosphere were measured. Based on the principle of differential optical absorption spectroscopy (DOAS) and the pressure calibration curve got in ~590 Torr nitrogen, we obtained the ammonia partial pressures. The result is compared with the value measured by vacuum gauge and the maximum error is 30%. This indicates that THz TDS based on principle of DOAS is an applicable quantitative technique for sensing ammonia or other polar gases in atmosphere.

We have conducted visible pump-THz probe experiments on single wall carbon nanotubes (SWCNTs) deposited on quartz substrates. Our results suggest that the photoexcited nanotubes exhibit localized transport due to Lorentz-type photo-induced localized states. These experiments were repeated for ion-implanted, 3-4nm Si nanoclusters in quartz for which a similar behavior was observed.

This paper presents the concept of an integrated millimeter-wave Fourier transform spectrometer. It is proposed to integrate the interferometer through the use of planar transmission line circuits onto a single chip. The splitting and combining of the broadband signal can be realized through the use of coupled-line couplers and the sliding reflector can be realized with distributed RF-MEMS transmission lines. Calculations are given to demonstrate the feasibility of this concept.

Pacific Advanced Technology has developed a small hand held imaging spectrometer, Sherlock, for gas leak and aerosol detection and imaging. The system is based on a patent technique that uses diffractive optics and image processing algorithms to detect spectral information about objects in the scene of the camera (IMSS Image Multi-spectral Sensing). This camera has been tested at Dugway Proving Ground and Dstl Porton Down facility looking at Chemical and Biological agent simulants. The camera has been used to investigate surfaces contaminated with chemical agent simulants. In addition to Chemical and Biological detection the camera has been used for environmental monitoring of green house gases and is currently undergoing extensive laboratory and field testing by the Gas Technology Institute, British Petroleum and Shell Oil for applications for gas leak detection and repair. The camera contains an embedded Power PC and a real time image processor for performing image processing algorithms to assist in the detection and identification of gas phase species in real time. In this paper we will present an over view of the technology and show how it has performed for different applications, such as gas leak detection, surface contamination, remote sensing and surveillance applications. In addition a sampling of the results form TRE field testing at Dugway in July of 2002 and Dstl at Porton Down in September of 2002 will be given.

This paper describes the evolution and capabilities of our Fourier transform infrared (FTIR) modulator in multiple configurations for various applications concerning Homeland Security. The heart of our system is the extremely compact and rugged Michelson interferometer we originally developed during our involvement with the JSLSCAD program. (1) Our "J-Series" modulator is capable of resolving the 7 to 14 μm spectral region to 4 or 16 cm^-1 with a measured radiometric sensitivity of 2x10^-9 W/(cm²ster·cm^-1). This system has successfully undergone rigorous testing for operation over a temperature range of -40 to +65°C and vibration levels associated with a spectrum of military ground, air, and water vehicles. The following describes the design and characterization of our J-Series modulator as well as the subsequent evolutions of the instrument in the forms of an active open-path FTIR and an imaging (hyperspectral) FTIR. We present this system in passive and active configurations with cooled, uncooled, and imaging detectors. We also project sensitivity limits for each configuration and measurement for some common chemical agents as well as industrial compounds.

The US Army Research Laboratory has developed a no-moving-parts, random wavelength access hyperspectral images using an electronically tunable acousto-optic tunable filter (AOTF) as a programmable dispersive element in combination with a camera and appropriate optics covering the wavelength region from the wavelength region from the ultraviolet (UV) to the visible (VIS). The imager operates from 220 to 480 nm with a spectral resolution of 160 cm^-1/ Sicj an UV imager can be used for a variety of chem/bio detection applications including monitoring of ozone in the atmosphere as well as in fluorescence microscopy to differentiate between normal and cancerous tissues for medical diagnosis. A novel large aperture AOTF was fabricated for this imager using a large single crystal of potassium dihydrophosphate (KH₂PO₄) commonly known as KPD. In the hyperspectral imager this AOTF is used with an uncooled enhanced charge couple device camera (CCD). In this paper, we will describe the AOTF cell, the imager, and present results from our imaging experiments.

Candidate weapon systems have conservative environmental and service life limits to ensure both performance reliability and ordnance safety. One important element that must be monitored is chemical indicators of propellant degradation. Chemical degradation of energetic compounds in propellants can result in reduced performance and potential instability and auto-ignition in extreme circumstances. The current method for testing for chemical indicators of propellant degradation consists of removing a missile from its sub, disassembling it, and performing HPLC testing. An improvement to the current system is to use near-infrared (NIR) spectral analysis to measure chemical indicators of propellant degradation. An AOTF multi-channel spectrometer with reflectance probes can simultaneously scan different areas of a propellant. A study has shown clear spectral differences in samples of M1MP propellant with two different concentrations of the chemical diphenyl amine (DPA). DPA is very similar to many important chemical indicators of propellant degradation. The spectral differences provide the basis for correlating spectral data to DPA concentration using a multivariate regression technique.

In this paper, the development of an active, wavelength agile, all-solid-state long-wave IR (8 - 12 μm) spectrapolarimetric imager for extraction of Stokes vector is discussed. Results obtained in regard to the development of a Hg₂Cl₂ based acousto-optic tunable filter and 1543 nm pumped AgGaSe₂ optical parametric oscillator are presented.

Pacific Northwest National Laboratory (PNNL) has recently developed a hybrid infrared technique for standoff chemical detection. Active infrared detection typically involves a sender and receiver telescope separated by (100’s) of meters and is quite sensitive, but is extremely cumbersome to align and is extremely sensitive to misalignment as the two telescopes must not only be parallel, but coaxial. Passive infrared sensing offers facile alignment (simply point the input optics), but relies on a happenstance temperature difference ΔT between the chemical plume and its background. Oftentimes the ΔT found in the field is only 1 or 2 K, and the passive method is thus not very sensitive in many cases. The “semi-active” technique creates a large temperature difference ΔT by placing an extended blackbody source at some distance away from the receiver telescope. The blackbody is designed to fill the telescope’s FOV at a typical distance of 100 m, and provides a typical temperature difference ΔT on the order of 80 to 100 K. Design considerations and experimental results in a direct comparison of passive, active, and semi-active measurements will be discussed.

Manning Applied Technology is developing a series of compact and rugged interferometers with many applications, including chemical and biological agent detection. A nutating prism interferometer features high throughput, very rapid-scan capability and resistance to vibration. A novel multiplex Fabry-Perot interferometric spectrometer is especially compact and efficient, transferring about twice the modulated optical power to a detector, compared to a Michelson interferometer of equal volume. The bilithic interferometer features high throughput, intrinsic tilt compensation and compact size with the potential for high resolution measurements. For all of these interferometer geometries, digital signal processing (DSP) hardware and software are used to enhance the performance and allow rapid processing of data, including chemometric discrimination. Enhancements to system performance provided by DSP include application of Brault's data processing approach, frequency stabilization of a miniature solid state reference laser and computational correction of tilt errors. A key advantage of the DSP enhancements is compensation for non-ideal behavior of optical hardware. Many of the novel instruments described herein employ intrinsic optic tilt compensation. Either type of tilt compensation provides greater photometric accuracy and stability. Preliminary NESR characterization is reported for two of the instruments.

In this paper, the development of a compact, electronically tunable liquid crystal tunable retarder (LCTR) for manipulation of polarization is presented. A LCTR with one-inch aperture diameter has been designed and fabricated for operation over mid-wave IR (MWIR) spectral band. Using this device, quarter and half-wave retardation have been demonstrated at several prominent laser wavelengths in the MWIR spectral region. Up to 74% efficiency has been demonstrated at 2.4 μm wavelength. Highly efficient retarders are now being developed using perfluorinated liquid crystal mixtures. An LCTR is anticipated to guide the development of a compact, wideband, and electronically tunable spectrapolarimetric imager.

A high repetition rate, wavelength agile CO₂ laser has been developed at the Air Force Research Laboratory for use as a local oscillator in a heterodyne detection receiver. Rapid wavelength selection is required for measurements of airborne chemical vapors using the differential absorption lidar (DIAL) technique. Acousto-optic modulators are used in the local oscillator to tune between different wavelengths at high speeds (greater than 100 Hz) without the need for moving mechanical parts. Other advantages obtained by the use of acousto-optic modulators are laser output power control per wavelength and rugged packaging for field applications. A series of experiments to simultaneously characterize the radiometric and chemical detection sensitivities of heterodyne and direct detection DIAL systems is being performed at Kirtland AFB, NM, and will be described. The wavelength agile local oscillator (WALO) has been incorporated into a heterodyne receiver, with the Laser Airborne Remote Sensing (LARS) system providing the laser transmitter and direct detection receiver. The experiment series is studying radiometric issues, spread spectrum operation, the effects of target-induced speckle, and the influence of atmospheric turbulence for both detection mechanisms. Measurements are being performed over a horizontal path at standoff ranges of 4 to 15 km, using both natural and man-made targets. Comparisons of the heterodyne and direct detection radiometric and chemometric results will be presented, and contrasted with predictions from simulations and models. The results will also be discussed in terms of the implications for fielding operational DIAL systems.

A compact long-wavelength infrared (LWIR) source, based on the combination of a high-pulse-rate, 1-μm-wavelength solid-state laser with rapid and broadly tunable tandem optical parametric oscillators (OPOs) was demonstrated. Nanosecond pulses with up to 100 μJ energy, tunable within the 8-11 μm LWIR range were achieved.

A study on the passive standoff detection and identification of Bacillus subtilis (BG) clouds with the Compact ATmospheric Sounding Interferometer (CATSI) sensor is presented. The analysis is based on recent spectral measurements of BG clouds obtained during the excursion phase of the Technology Readiness Evaluation trial held at Dugway Proving Ground, 15-26 July 2002. Good results from two trial episodes corresponding to the measurement of BG clouds at a distance of 3 km in a near-horizontal path scenario are used to explain and demonstrate the detection capability of the CATSI sensor. It has been found that the low thermal contrast (few tenths of a degree) between the BG cloud and the background yields weak but observable spectral signatures. The results of a series of simulations with the FASCODE transmission model have shown that the detection sensitivity for BG can be greatly improved for slant path scenarios.

Outdoor measurements of dry bacillus subtilis (BG) spores were conducted with a passive Fourier transform infrared (FTIR) spectrometer using two types of chambers. One was a large open-ended cell, and the other was a canyon of similar dimensions. The canyon exposes the aerosol plume to downwelling sky radiance, while the open-ended cell does not. The goal of the experiments was to develop a suitable test methodology for evaluation of passive standoff detectors for open-air aerosol measurements. Dry BG aerosol particles were dispersed with a blower through an opening in the side of the chamber to create a pseudo-stationary plume, wind conditions permitting. Numerous trials were performed with the FTIR spectrometer positioned to view mountain, sky and mixed mountain-sky backgrounds. This paper will discuss the results of the FTIR measurements for BG and Kaolin dust releases.

A novel windowless chamber was developed to allow aerosol backscatter measurements with a frequency-agile CO₂ lidar. The chamber utilizes curtains of air to contain the cloud, thus preventing the inevitable backscatter off of conventional windows from corrupting the desired measurements. This feature is critical because the CO₂ lidar has a long (1 μs) pulse and the backscatter off the window cannot be temporally separated from the backscatter off the aerosol in the chamber. The chamber was designed for testing with a variety of CB simulants and interferents in both vapor and aerosol form and has been successfully shown to contain a cloud of known size, concentration, and particle size distribution for 10-15 minutes. This paper shows the results using Arizona road dust that was screened by the manufacturer into 0-3 μm and 5-10 μm particle size distributions. The measurements clearly show the effect of size distribution on the infrared backscatter coefficients as well as the dynamic nature of the size distribution for a population of aerosols. The test methodology and experimental results are presented.

The Airborne Chemical Imaging System (ACIS) is a research platform used to evaluate passive infrared (IR) standoff detectors for airborne remote sensing of chemical vapors. It consists of a sensor suite mounted in an automated gyro-stabilized optical platform. The sensor pod is currently mounted on a UH-1 helicopter but could also be adapted to other platforms. Two developmental IR imaging sensors are used in the ACIS: a high-speed Fourier transform infrared (FTIR) spectrometer: the TurboFT, and a high-resolution tunable IR Fabry-Perot spectroradiometer: the AIRIS. The TurboFT is a high-speed (100 Hz) low-resolution (2x8 pixel) system and the AIRIS is a low-speed (~0.5 Hz), high-resolution (64x64 pixel) imager. This paper describes the ACIS configuration, general system specifications, operational concerns, and some typical results from recent flight tests.

Airborne testing of sensors presents unique challenges to the researcher. Prototype sensors are not typically configured for aircraft mounting, and testing requires comparative (truth) data for accurate sensor performance evaluation. The U.S. Army Redstone Technical Test Center (RTTC) has developed a large Stabilized Electro-optical Airborne Instrumentation Platform (SEAIP) for use with rotary wing aircraft as a sensor test bed. This system is designed to accommodate the rapid integration of multiple sensors into the gimbal, greatly reducing the time required to enter a sensor into testing. The SEAIP has been designed for use with UH-1 or UH-60 aircraft. It provides nominal 35 μradian (RMS) line-of-sight stabilization in two axes. Design has been optimized for support of multiple/large prototype (brassboard) sensors. Payload combinations up to 80 lbs can be accommodated. Gimbal angle ranges are large to permit flexibility for sensor pointing. Target acquisition may be done manually, or with the use of a GPS tracker. Non-visible targets may be engaged, and sensor information may be mapped real-time to digitized maps or photographs of the test area. Two SEAIP systems are currently used at RTTC. Numerous sensors have been successfully integrated and tested, including MMW, LADAR, IR, SAL, multi-spectral, visible, and night vision.

An investigation is made into the possibility of applying the passive standoff detection technique to the identification of radiological and related products. This work is based on laboratory measurements of the diffuse reflectance from a number of radiological or related products, including U₃O₈, ThO₂, CsI, SrO, I₂O₅ and La₂O₃. With the use of these measured reflectances, simulations of the nadir radiances with the various types of surface reflectances were carried out with the MODTRAN4 transmission model. The simulations were performed for two types of scenarios; at an altitude of 1 m above the ground for the purpose of simulating the passive detection of nuclear products with a hand-held instrument, and at an altitude of 1 km to emulate the conditions of a passive sensor carried aloft in an aircraft. The results of the simulations under idealized conditions show that there is a good potential for being able to measure radiological products or related materials by passive standoff detection using Fourier-transform infrared radiometric techniques.

A technique for the remote sensing of the forcing function of global warming (i.e., the increase in the surface radiative forcing) is described. Climate models predict that the emission of greenhouse gases into the atmosphere has altered the radiative energy balance at the earth's surface through increasing the greenhouse radiation from the atmosphere. With measurements at high spectral resolution, this increase can be unambiguously attributed to each of several anthropogenic gases. Calibrated spectra of the greenhouse radiation from the atmosphere have been measured at ground level from Peterborough, Ontario using an FTIR spectrometer with a resolution of 0.1 cm^-1. This long-wave radiation consists of the thermal emission from naturally occurring gases, such as CO₂, H₂O and O₃, as well as from many trace gases, such as CH₄, CFC11, CFC12, CFC22 and HNO₃. The forcing radiative fluxes from CFC11, CFC12, CCl₄, N₂O, CH₄, HNO₃ and CO₂ have been quantitatively measured. Measurements of the fluxes associated with each gas are presented. Models indicate that an energy flux imbalance of about 3 W/m² has been created by anthropogenic emissions of greenhouse gases of which we have measured over 1.0 W/m². Much of the remaining flux change is due to CO₂ and CH₄ increases which are difficult to measure without a long baseline data series. Overall, it has been demonstrated that the greenhouse radiation is measurable and it should be monitored on a worldwide basis over the long term since the predicted increase in this energy flux is the fundamental forcing of global warming.

The SciSat-1 mission is a dedicated Canadian science satellite that will investigate processes that control the distribution of ozone in the stratosphere. The SciSat-1 satellite consists of primarily two science instruments; an Atmospheric Chemistry Experiment (ACE) high-resolution Fourier-transform spectrometer (FTS) and an ultraviolet-visible-near-infrared spectrograph. These instruments will primarily function in occultation mode; however, during the dark portion of the orbit the Earth will pass between the sun and the satellite. This configuration will give rise to the opportunity of acquiring some nadir-view FTIR spectra of the Earth. Since the ACE FTS was designed to view a hot source (i.e., the Sun) at high resolution using a single scan, it is necessary to determine if the FTS will provide nadir spectra of the relatively cold atmosphere and surface with a sufficient signal-to-noise ratio. Methane, ozone and carbon monoxide gases were used in the cell for the purpose of determining the measurement characteristics of the ACE FTS instrument for a low-intensity source. These measurements were compared with data obtained from the Interferometric Monitor for Greenhouse (IMG) gases onboard the ADEOS satellite. The results show that the ACE FTS should be able to measure the abundant trace gases in the atmosphere with sufficient signal-to-noise ratio.

Standoff detection of chemical agents may be enhanced with the capability to measure an image of the agent concentration. The use of an imaging Fourier-Transform Spectrometer to perform these measurements is extensively modeled in order to predict its ultimate capabilities. The model developed allows one to determine the optimal configuration of the instrument, taking into account the precise characteristics of realistic and existing hardware. The model is first based on the calculation of radiative transfer from the scene into the instrument up to the imaging detector. Standard performance models of FTS are improved to include the particular features of imaging FTS operated with infrared cameras. The infrared focal plane arrays have their own constraints that are taken into account in the model.

Spectral signature coding is an effective means to characterize spectral features. This paper presents a novel progressive spectral signature coding, called Multistage Predictive Coding Modulation (MPCM) for hyperspectral imagery. It not only can be used for signature coding, but also can provide a progressive profile for each signature in terms of identification. The idea is to consider the spectrum of a hyperspectral image pixel vector as a 1-D signal, then the proposed MPCM encodes such a 1-D spectral signal progressively based on the priorities assigned to signal points for spectral reconstruction. The MPCM generates priority codewords for a spectral signature in a specific manner that the characteristics of this particular signature can be captured and used as its own fingerprint. Despite that the success of MPCM in spatial coding has been demonstrate in image progressive reconstruction, progressive edge detection and text detection, its utility in spectral coding has not been investigated and is yet to be explored. This paper demonstrates that the MPCM-based progressive spectral coding can be useful not o nly in data compression, progressive reconstruction, but also in progressive material identification.

The Constrained Energy Minimization (CEM) has been widely used for hyperspectral detection and classification. The feasibility of implementing the CEM as a real-time processing algorithm in systolic arrays has been also demonstrated. The main challenge of realizing the CEM in hardware architecture in the computation of the inverse of the data correlation matrix performed in the CEM, which requires a complete set of data samples. In order to cope with this problem, the data correlation matrix must be calculated in a causal manner which only needs data samples up to the sample at the time it is processed. This paper presents a Field Programmable Gate Arrays (FPGA) design of such a causal CEM. The main feature of the proposed FPGA design is to use the Coordinate Rotation DIgital Computer (CORDIC) algorithm that can convert a Givens rotation of a vector to a set of shift-add operations. As a result, the CORDIC algorithm can be easily implemented in hardware architecture, therefore in FPGA. Since the computation of the inverse of the data correlction involves a series of Givens rotations, the utility of the CORDIC algorithm allows the causal CEM to perform real-time processing in FPGA. In this paper, an FPGA implementation of the causal CEM will be studied and its detailed architecture will be also described.

Hyperspectral images have high spectral resolution that helps to improve object classification. But its vast data volume also causes problems in data transmission and data storage. Since there is high correlation among spectral bands in a hyperspectral image, how to reduce the data redundancy while keeping the important information for the following data analysis is a challenging task. In this paper, we investigate a compression technique based on segmented Principal Components Analysis (PCA). A hyperspectral image cube is divided into several non-overlapping blocks in accordance with band-to-band cross-correlations, followed by the PCA performed in each block. A major advantage resulting from this approach is computational efficiency. The utility of the proposed segmented PCA-based compression in target dtection and classification will be investigated. The experiments demonstrate that the segmented PCA-based compression generally outperforms PCA-based compression in terms of high detection and classification accuracy on decompressed hyperspectral image data.

Hyperspectral imaging has recently received considerable interest in land-cover classification. With the improvement of spectral resolution, hyperspectral images can be used to detect and classify subtle land cover types which cannot be resolved by multispectral data. Unfortunately, most of techniques for land cover classification are developed based on pattern classification rather than target classification. The chief difference between these two is that patter classification is performed by classifying all image pixel vectors into different types of pattern classes, including image background, whereas target classification is conducted based on targets of interest regardless of what image background is. This paper presents hyperspectral land-cover classification techniques based on targets of interest. Experiments are conducted using DAIS data acquired by GER for applications in agriculture and environmental monitoring.

A mixture model for hyperspectral data describes the underlying physics of how constituent substances in a scene affect each other to produce radiance received at each detector pixel. In the linear models the total scene is portrayed as a checkerboard mixture of reflecting surfaces containing fractional abundances of the constituent substances. The radiance measured at a detector pixel is modeled proportional to the same abundance of the constituents wehre interference (interaction) between constituent substances is neglected. In this paper we present a model to account for a multiple reflection process wehreby the measured radiation is the result of multiple inelastic interactions (scattering or reflection) among the constituent components in the scene. While this process is inherently non-linear, and we have constructed an equivalent "extended" linear version of the model, which yields the effective abundance of the end members in the scene. This model seamlessly reduces to the conventional linear model when multiple scattering is absent. We present results of preliminary applications of these models with standard unmixing algorithm. The results show that the inclusion of higher order reflections always produces progressively a better description of the abundances and the interactions between the endmembers in the scene.

Laboratory and field measurements for the passive standoff detection of liquid contaminants deposited on surfaces are presented. The measurements were performed at the SURFCON trials held at DRDC Suffield in September 2002. The goal of this trial was to verify that passive LWIR spectrometric sensors have the potential for remotely detecting surface contaminants. Laboratory and field data obtained with the passive CATSI sensor are analyzed. For laboratory measurements, variable amounts of liquid H-contaminant were deposited on high-reflectivity and low-reflectivity surfaces of aluminum and Mylar. Field measurements were performed at a standoff distance of 60 m on aluminium plates covered with the H-contaminant and on low-reflectivity surfaces (armored personnel carrier surface and sod). Results indicate that contaminants deposited on high-reflectivity surfaces can be detected, identified and possibly quantified. The contaminants deposited on low-reflectivity surfaces can usually be detected but cannot be identified based on simple correlations with the absorption spectrum of the contaminant. The inhomogeneous layer model developed to explain the spectral radiance associated with non-uniform contaminant coverage is also presented.

Laser Interrogation of Surface Agents (LISA) is a UV-Raman technique that provides short-range standoff detection and identification of surface-deposited chemical agents. ITT Industries, Advanced Engineering and Sciences Division, is currently developing and expanding the LISA technology under several programs that span a variety of missions for homeland defense. We will present and discuss some of these applications, while putting in perspective the overall evolution undergone by the technique within the last years. These applications include LISA-Recon (now called the Joint Contaminated Surface Detector--JCSD) which was developed under a cost-sharing arrangement with the U.S. Army Soldier and Biological Chemical Command (SBCCOM) for incorporation on the Army’s future reconnaissance vehicles, and designed to demonstrate single-shot on-the-move measurements of chemical contaminants at concentration levels below the Army's requirements. In parallel, LISA-Shipboard is being developed to optimize the sensor technique for detection of surface contaminants in the operational environment of a ship. The most recently started activity is LISA-Inspector that is being developed to provide a transportable sensor in a 'cart-like' configuration.

We present our work towards developing a compact reflector telescope (CRT) for short-range (1 to 50 m) standoff Raman LIDAR applications, including a standoff Raman measurement employing our telescope wiht a commercial off the shelf (COTS) laser, spectrometer, and Raman edge filter. This development effort was funded through an SBIR contract from the Department of Energy. The application of this technology is standoff assessment of chemical spills. The CRT system includes a small Galilean telescope to deliver the excitation beam to the surface under investigation; the benefit of the delivery optics is a smaller laser spot at the target and significantly enhanced throughput relative to systems which rely on the divergence of the excitation laser beam. The CRT itself is a 10-inch Cassegrain optimized for this short standoff range with motor-driven focus adjustment. We executed a Raman measurement of acetone at a standoff of 2 m using a Midwest Laser 325 nm helium cadmium laser, an Ocean Optics USB2000 grating spectrometer (with uncooled CCD), and an Omega edge filter. We present the results overlayed with published reference spectra. To the best of our knowledge, this is the first reported standoff Raman measurement performed with an uncooled CCD detector.

Protection of the drinking water supply from a terrorist attack is of critical importance. Since the water supply is vast, contamination prevention is difficult. Therefore, rapid detection of contaminants, whether a military chemical/biological threat, a hazardous chemical spill, naturally occurring toxins, or bacterial build-up is a priority. The development of rapid environmentally portable and stable monitors that allow continuous monitoring of the water supply is ideal. EIC Laboratories has been developing Surface-Enhanced Raman Spectroscopy (SERS) to detect chemical agents, toxic industrial chemicals (TICs), viruses, cyanotoxins and bacterial agents. SERS is an ideal technique for the Joint Service Agent Water Monitor (JSAWM). SERS uses the enhanced Raman signals observed when an analyte adsorbs to a roughened metal substrate to enable trace detection. Proper development of the metal substrate will optimize the sensitivity and selectivity towards the analytes of interest.

The motivation of this work is to monitor the real time concentration of nitrates in the radioactive wastes, as they are the key molecules in the solution. The effect of different optical configurations of the probes on the Raman signal was studied using acetone as the test sample. We found that InPhotonics^TM Raman probe give best signal-to-noise data comparing to the other two probes evaluated. The Raman spectra of 10% NaNO₃ solution were then successfully recorded with this probe. The Raman signal of Nitrate at 1054 cm^-1 is very strong with 500 ms sampling time. The initial study shows that the Raman sensor is capable to monitor the nitrate in the nuclear waste tank.

This is an overview of the work carried out in the UK on the stand-off detection of liquid contamination. The UK uses a two-stage concept employing LWIR (long wave infrared) reflectance imaging for location followed by Laser Induced Vapour Emission (LIVE) (patent pending) for identification of the material. Research has been conducted into IR reflectance imaging, using a HgCdTe starring array and broad band source. 2-5mm diameter contaminant droplets were resolved at distances of 5.5 m on both painted plates and asphalt. Short distances LIVE experiments using CW agents produced characteristic LWIR emission spectra. These spectra show clear differences between VX, GD and HD as well as backgrounds such as oil and water. Droplets were found to vaporise more efficiently from less absorbent surfaces such as metal and asphalt. A pyroprobe (a rapidly heating probe) was used to flash heat 20 μl droplets of HD, which was detected at 1 metre in a previous version of the experiment. Longer distance experiments were successfully carried out using smaller amounts of simulant at distances of 18 m. This suggests identification of agent at 20 metres should be trivial providing the rapid heating and generation of hot vapour by remote means is successful. Further, the method is rapid: time resolved studies using a spectroradiometer capable of producing 20 spectra/second shows that 1 second data acquisition is sufficient.