Abstract: This disclosure relates to processing a spectral dataset, such as a hyperspectral image or a large collection of individual spectra taken with the same spectrometer, to increase the signal-to-noise ratio. The methods can also be used to process a stack of images that differ by acquisition time rather than wavelength. The methods remove most of the sensor background noise with minimal corruption of image texture, anomalous or rare spectra or waveforms, and spectral or time resolution.

Abstract: This disclosure relates to processing a spectral dataset, such as a hyperspectral image or a large collection of individual spectra taken with the same spectrometer, to increase the signal-to-noise ratio. The methods can also be used to process a stack of images that differ by acquisition time rather than wavelength. The methods remove most of the sensor background noise with minimal corruption of image texture, anomalous or rare spectra or waveforms, and spectral or time resolution.

Abstract: The invention provides a method for identifying one or more materials in a scene by determining a set of spectral vectors, called endmembers, from a data set comprised of spectra from the image data, and matching the set of endmembers to predefined library materials. The image data of the scene is captured with a sensor, and comprises a plurality of spectra. The method applies an iterative mathematical criterion, termed residual minimization, to find the endmembers. The first endmember may be selected based on the largest mean square value or the largest mean magnitude value. Subsequent endmembers are determined by calculating weighting factors, such that the weighting factors are non-negative and the calculated vector differences, or residuals, generate the smallest error metric. The error metric is dependent upon the vector difference between two spectra in the image data set, and may be the mean squared vector difference between two spectra.

Abstract: A spectral encoder for producing spectrally selected images of a radiation field containing multiple spectral components. An imaging spectrograph defines a first optical path that produces from the input radiation field a spectrally dispersed image comprising multiple spectral components displaced along a dispersion direction. Spectral pass bands are encoded on the dispersed image by a programmable spatial light modulator using one or more spatial masks. The imaging spectrograph further defines a second optical path that reverses the spectral dispersion of the first path and produces a spectrally-encoded polychromatic output image containing only those spectral components encoded by the spatial mask. The first and second optical paths share a common dispersing element. A detector records at least one spatial region of the spectrally encoded output image.

Abstract: A radiative transport band model method for prediction and analysis of high spectral resolution radiometric measurements. Atomic and molecular line center absorption is determined from finite spectral bin equivalent widths. A mathematically exact expansion for finite bin equivalent widths provides high accuracy at any desired spectral resolution. The temperature and pressure dependent Voigt line tail spectral absorption contributing to each spectral bin is pre-computed and fit to Padé approximants for rapid and accurate accounting of neighboring-to-distant lines. A specific embodiment has been incorporated into the MODTRAN™ radiation transport model.

Abstract: A method of representing spectral data, such as hyperspectral imaging data (HSI) and multispectral imaging data (MSI), as a set of simplex models. The method finds end-images or end-spectra in the data (termed “endmembers”) as extreme points, and simultaneously determines the abundance of the endmembers.

Abstract: A method of automatically determining a measure of atmospheric aerosol optical properties using a multi- or hyper-spectral, multi-pixel image. A plurality of spectrally-diverse pixels are resolved from the image. A statistical spectral deviation of the spectrally-diverse pixels is determined, and then corrected for non-aerosol transmittance losses. One or more wavelength-dependent aerosol optical depths are derived from the statistical spectral deviation. Wavelength-dependent gaseous optical depths can be derived from the statistical spectral deviation.

Abstract: This invention discloses several improved methods of correcting for atmospheric effects on a remote image of the Earth's surface taken from above, wherein the image comprises a number of simultaneously acquired images of the same scene, each including a large number of pixels, each at a different wavelength band, and including infrared through ultraviolet wavelengths. One method is for retrieving the aerosol/haze amount (i.e., visible range) from an assumed ratio of in-band reflectances, rather than from an assumed reflectance value. Another method is for identifying cloud-containing pixels. This is used to improve the calculation of the spatially averaged radiance L*e and reflectance &rgr;e images in standard equations. Another method greatly reduces the number of mathematical operations required to generate the reflectance values.

Abstract: A system and method for optical interrogation and measurement of a hydrocarbon fuel gas includes a light source generating light at near-visible wavelengths. A cell containing the gas is optically coupled to the light source which is in turn partially transmitted by the sample. A spectrometer disperses the transmitted light and captures an image thereof. The image is captured by a low-cost silicon-based two-dimensional CCD array. The captured spectral image is processed by electronics for determining energy or BTU content and composition of the gas. The innovative optical approach provides a relatively inexpensive, durable, maintenance-free sensor and method which is reliable in the field and relatively simple to calibrate. In view of the above, accurate monitoring is possible at a plurality of locations along the distribution chain leading to more efficient distribution.

Abstract: An ambient surface contamination sensor including a source for supplying gas, which is energized before it reaches the surface being sensed so that it is capable of transferring energy to the contaminants on the surface or vapor originating therefrom. The sensor then detects the optical emission from the gas passed over the surface within a selected wavelength band characteristic of the presence on the surface of the contaminant, and indicates the presence of the contaminant on the surface when such an emission is detected.

Abstract: A gas species monitor system includes a sample volume for receiving a gas to be monitored; an external, independent laser source; means for directing the laser radiation to the volume; a multipass optical cell, responsive to the means for directing, for multiplying the laser radiation intensity in the sample volume; means for continuously flowing the gas to be monitored through the sample volume; a narrow bandpass filter; means for collecting more than a steradian of Raman scattered radiation from the sample in the volume and directing the collected radiation in parallel through the narrow bandpass filter; and means responsive to the parallel radiation from the filter for detecting the Raman scattered radiation representative of the concentration of the species in the gas sample being monitored.

Abstract: A spark discharge trace element detection system is provided which includes a spark chamber including a pair of electrodes for receiving a sample of gas to be analyzed at no greater than atmospheric pressure. A voltage is provided across the electrodes for generating a spark in the sample. The intensity of the emitted radiation in at least one primary selected narrow band of the radiation is detected. Each primary band corresponds to an element to be detected in the gas. The intensity of the emission in each detected primary band is integrated during the afterglow time interval of the spark emission and a signal representative of the integrated intensity of the emission in each selected primary bond is utilized to determine the concentration of the corresponding element in the gas.

Abstract: A spark spectroscopic high-pressure gas analyzer including a spark chamber, having a pair of electrodes, for receiving a sample of the pressurized gas to be analyzed. A voltage is provided across the electrodes for generating a spark in the pressurized gas sample. A selected wavelength band of radiation emitted from the spark discharge in the pressurized gas corresponding to a component to be sensed in the gas is detected. The intensity of the emission in the wavelength band is integrated during the afterglow time interval of the spark emission and a signal representative of the integrated intensity of the emission in the selected narrow wavelength band is employed to determine the proportion of the component in the gas.