Steven W. Brown

Simple Spectral Stray Light Correction Method for Array Spectroradiometers

A simple, practical method has been developed to correct a spectroradiometers response for measurement errors arising from the instruments spectral stray light. By characterizing the instruments response to a set of monochromatic laser sources that cover the instruments spectral range, one obtains a spectral stray light signal distribution matrix that quantifies the magnitude of the spectral stray light signal within the instrument. By use of these data, a spectral stray light correction matrix is derived and the instruments response can be corrected with a simple matrix multiplication. The method has been implemented and validated with a commercial CCD-array spectrograph. Spectral stray light errors after the correction was applied were reduced by 1-2 orders of magnitude to a level of approximately 10(-5) for a broadband source measurement, equivalent to less than one count of the 15-bit-resolution instrument. This method is fast enough to be integrated into an instruments software to perform real-time corrections with minimal effect on acquisition speed. Using instruments that have been corrected for spectral stray light, we expect significant reductions in overall measurement uncertainties in many applications in which spectrometers are commonly used, including radiometry, colorimetry, photometry, and biotechnology.

Facility for spectral irradiance and radiance responsivity calibrations using uniform sources

Detectors have historically been calibrated for spectral power responsivity at the National Institute of Standards and Technology by using a lamp-monochromator system to tune the wavelength of the excitation source. Silicon detectors can be calibrated in the visible spectral region with combined standard uncertainties at the 0.1% level. However, uncertainties increase dramatically when measuring an instruments spectral irradiance or radiance responsivity. We describe what we believe to be a new laser-based facility for spectral irradiance and radiance responsivity calibrations using uniform sources (SIRCUS) that was developed to calibrate instruments directly in irradiance or radiance mode with uncertainties approaching or exceeding those available for spectral power responsivity calibrations. In SIRCUS, the emission from high-power, tunable lasers is introduced into an integrating sphere using optical fibers, producing uniform, quasi-Lambertian, high-radiant-flux sources. Reference standard irradiance detectors, calibrated directly against national primary standards for spectral power responsivity and aperture area measurement, are used to determine the irradiance at a reference plane. Knowing the measurement geometry, the source radiance can be readily determined as well. The radiometric properties of the SIRCUS source coupled with state-of-the-art transfer standard radiometers whose responses are directly traceable to primary national radiometric scales result in typical combined standard uncertainties in irradiance and radiance responsivity calibrations of less than 0.1%. The details of the facility and its effect on primary national radiometric scales are discussed.

NIST facility for Spectral Irradiance and Radiance Responsivity Calibrations with Uniform Sources

A laser-based facility has been developed to provide high-flux, monochromatic, Lambertian radiation over the spectral range 0.2 µm to 18 µm. The facility was designed to reduce the uncertainties in a variety of radiometric applications, including irradiance and radiance responsivity calibrations. The operational characteristics of the facility are discussed and the results of detector responsivity calibrations over the spectral range 0.406 µm to 0.920 µm are presented.

LED-based Spectrally Tunable Source for Radiometric, Photometric, and Colorimetric Applications

A spectrally tunable light source using a large number of LEDs and an integrating sphere has been designed and is being constructed at the National Institute of Standards and Technology. The source is designed to have a capability of producing any spectral distribution, mimicking various light sources in the visible region by feedback control of individual LEDs. The output spectral irradiance or radiance of the source will be calibrated by a reference instrument, and the source will be used as a spectroradiometric as well as a photometric and colorimetric standard. A series of simulations have been conducted to predict the performance of the designed tunable source when used for calibration of display colorimeters. The results indicate that the errors can be reduced by an order of magnitude when the tunable source is used to calibrate the colorimeters, compared with measurement errors when the colorimeters are calibrated against Illuminant A. The source can also approximate various CIE daylight illuminants and common lamp spectral distributions for other photometric and colorimetric applications.

Overview of the Radiometric Calibration of MOBY

The Marine Optical Buoy (MOBY) provides values of water- leaving radiance for the calibration and validation of satellite ocean color instruments. Located in clear, deep ocean waters near the Hawaiian Island of Lanai, MOBY measures the upwelling radiance and downwelling irradiance at three levels below the ocean surface plus the incident solar irradiance just above the surface. The radiance standards for MOBY are two integrating spheres with calibrations based on standards traceable to the National Institute of Standards and Technology (NIST). For irradiance, the MOBY project uses standard lamps that are routinely calibrated at NIST. Wavelength calibrations are conducted with a series of emission lines observed from a set of low pressure lamps. Each MOBY instrument views these standards before and after its deployment to provide system responses (calibration coefficients). During each deployment, the stability of the MOBY spectrographs and internal optics are monitored using three internal reference sources. In addition, the collection optics for the instrument are cleaned and checked on a monthly basis while the buoy is deployed. Divers place lamps over the optics before and after each cleaning to monitor changes at the system level. As a hyperspectral instrument, MOBY uses absorption lines in the solar spectrum to monitor its wavelength stability. When logistically feasible during each deployment, coincident measurements are made with the predecessor buoy before that buoys recovery. Measurements of the underwater light fields from the deployment vessel are compared with those from the buoy. Based on this set of absolute calibrations and the suite of stability reference measurements, a calibration history is created for each buoy. These calibration histories link the measurement time series from the set of MOBY buoys. In general, the differences between the pre- and post-deployment radiance calibrations of the buoys range from +1% to -6% with a definitive bias to a negative difference for the post- deployment values. This trend is to be expected after a deployment of 3 months. To date, only the pre-deployment calibration measurements have been used to adjust the system responses for the MOBY time series. Based on these results, the estimated radiometric uncertainty for MOBY in-water ocean color measurements is estimated to be about 4% to 8% (kequals1). As part of a collaboration with NIST, annual radiometric comparisons are made at the MOBY calibration facility. NIST personnel use transfer radiometers and integrating spheres to validate (verify) the accuracy of the MOBY calibration sources. Recently, we began a study of the stray light contribution to the radiometric uncertainty in the MOBY systems. A complete reprocessing of the MOBY data set, including the changes within each MOBY deployment, will commence upon the completion of the stray light characterization, which is scheduled for the fall of 2001. It is anticipated that this reprocessing will reduce the overall radiometric uncertainty to less than 5% (kequals1).

Stray-Light Correction Algorithm for Spectrographs

In this paper, we describe an algorithm to correct a spectrographs response for stray light. Two recursion relations are developed:?one to correct the system response when measuring broad-band calibration sources, and a second to correct the response when measuring sources of unknown radiance. The algorithm requires a detailed understanding of the effect of stray light in the spectrograph on the instruments response. Using tunable laser sources, a dual spectrograph instrument designed to measure the up-welling radiance in the ocean was characterized for stray light. A?stray-light correction algorithm was developed, based on the results of these measurements. The instruments response was corrected for stray light, and the effects on measured up-welling in-water radiance were evaluated.

Development of a Tunable LED-Based Colorimetric Source.

A novel, spectrally tunable light-source utilizing light emitting diodes (LEDs) for radiometric, photometric, and colorimetric applications is described. The tunable source can simulate standard sources and can be used as a transfer source to propagate photometric and colorimetric scales from calibrated reference instruments to test artifacts with minimal increase in uncertainty. In this prototype source, 40 LEDs with 10 different spectral distributions were mounted onto an integrating sphere. A voltage-to-current control circuit was designed and implemented, enabling independent control of the current sent to each set of four LEDs. The LEDs have been characterized for stability and dependence on drive current. The prototype source demonstrates the feasibility of development of a spectrally tunable LED source using LEDs with up to 40 different spectral distributions. Simulations demonstrate that such a source would be able to approximate standard light-source distributions over the visible spectral range—from 380 nm to 780 nm—with deviations on the order of 2 %. The tunable LED source can also simulate spectral distributions of special sources such as discharge lamps and display monitors. With this tunable source, a test instrument can be rapidly calibrated against a variety of different source distributions tailored to the anticipated uses of the artifact. Target uncertainties for the calibration of test artifacts are less than 2 % in luminance and 0.002 in chromaticity for any source distribution.

Hyperspectral image projectors for radiometric applications

We describe a Calibrated Hyperspectral Image Projector (CHIP) intended for radiometric testing of instruments ranging from complex hyperspectral or multispectral imagers to simple filter radiometers. The CHIP, based on the same digital mirror arrays used in commercial Digital Light Processing (DLP) displays, is capable of projecting any combination of as many as approximately one hundred different arbitrarily programmable basis spectra per frame into each pixel of the instrument under test (IUT). The resulting spectral and spatial content of the image entering the IUT can simulate, at typical video frame rates and integration times, realistic scenes to which the IUT will be exposed during use, and its spectral radiance can be calibrated with a spectroradiometer. Use of such generated scenes in a controlled laboratory setting would alleviate expensive field testing, allow better separation of environmental effects from instrumental effects and enable system-level performance testing and validation of space-flight instruments prior to launch. Example applications are system-level testing of complex hyperspectral imaging instruments and algorithms with realistic scenes and testing the performance of first-responder cameras under simulated adverse conditions. We have built and tested a successful prototype of the spectral engine, a primary component of the CHIP, that generates arbitrary, programmable spectra in the 1000 nm to 2500 nm spectral range. We have also built a spectral engine operating at visible wavelengths to be discussed in a separate publication. Here we present an overview of this technology and its applications and discuss experimental performance results of our prototype infrared spectral engine.

Spectrally Tunable Sources for Advanced Radiometric Applications

A common radiometric platform for the development of application-specific metrics to quantify the performance of sensors and systems is described. Using this platform, sensor and system performance may be quantified in terms of the accuracy of measurements of standardized sets of source distributions. The prototype platform consists of spectrally programmable light sources that can generate complex spectral distributions in the ultraviolet, visible and short-wave infrared regions for radiometric, photometric and colorimetric applications. In essence, the programmable spectral source is a radiometric platform for advanced instrument characterization and calibration that can also serve as a basis for algorithm testing and instrument comparison.

A hyperspectral image projector for hyperspectral imagers

We have developed and demonstrated a Hyperspectral Image Projector (HIP) intended for system-level validation testing of hyperspectral imagers, including the instrument and any associated spectral unmixing algorithms. HIP, based on the same digital micromirror arrays used in commercial digital light processing (DLP*) displays, is capable of projecting any combination of many different arbitrarily programmable basis spectra into each image pixel at up to video frame rates. We use a scheme whereby one micromirror array is used to produce light having the spectra of endmembers (i.e. vegetation, water, minerals, etc.), and a second micromirror array, optically in series with the first, projects any combination of these arbitrarily-programmable spectra into the pixels of a 1024 x 768 element spatial image, thereby producing temporally-integrated images having spectrally mixed pixels. HIP goes beyond conventional DLP projectors in that each spatial pixel can have an arbitrary spectrum, not just arbitrary color. As such, the resulting spectral and spatial content of the projected image can simulate realistic scenes that a hyperspectral imager will measure during its use. Also, the spectral radiance of the projected scenes can be measured with a calibrated spectroradiometer, such that the spectral radiance projected into each pixel of the hyperspectral imager can be accurately known. Use of such projected scenes in a controlled laboratory setting would alleviate expensive field testing of instruments, allow better separation of environmental effects from instrument effects, and enable system-level performance testing and validation of hyperspectral imagers as used with analysis algorithms. For example, known mixtures of relevant endmember spectra could be projected into arbitrary spatial pixels in a hyperspectral imager, enabling tests of how well a full system, consisting of the instrument + calibration + analysis algorithm, performs in unmixing (i.e. de-convolving) the spectra in all pixels. We discuss here the performance of a visible prototype HIP. The technology is readily extendable to the ultraviolet and infrared spectral ranges, and the scenes can be static or dynamic.