Simulated Radiance Profiles for Automating the Interpretation of Airborne Passive Multi-Spectral Infrared Images

Abstract

Methodology is developed for simulating the radiance profiles acquired from airborne passive multispectral infrared imaging measurements of ground sources of volatile organic compounds (VOCs). The simulation model allows the superposition of pure-component laboratory spectra of VOCs onto spectral backgrounds that simulate those acquired during field measurements conducted with a downward-looking infrared line scanner mounted on an aircraft flying at an altitude of 2000-3000 ft (approximately 600-900 m). Wavelength selectivity in the line scanner is accomplished through the use of a multichannel Hg:Cd:Te detector with up to 16 integrated optical filters. These filters allow the detection of absorption and emission signatures of VOCs superimposed on the upwelling infrared background radiance within the instrumental field of view (FOV). By combining simulated radiance profiles containing analyte signatures with field-collected background signatures, supervised pattern recognition methods can be employed to train automated classifiers for use in detecting the signatures of VOCs during field measurements. The targeted application for this methodology is the use of the imaging system to detect releases of VOCs during emergency response scenarios. In the work described here, the simulation model is combined with piecewise linear discriminant analysis to build automated classifiers for detecting ethanol and methanol. Field data collected during controlled releases of ethanol, as well as during a methanol release from an industrial facility, are used to evaluate the methodology.

Threshold concentration monitoring based on pattern recognition analysis of differential near-infrared spectra

“…The PLDA method uses multiple linear discriminant functions to approximate a nonlinear separation boundary between two data classes. We have used this method in a number of applications in our laboratory, [20][21][22] and found it to be quick to implement in an automated manner without the requirement for extensive optimization of its architecture. Because the optimization of the discriminant functions is iterative in nature, three replicate classication models were constructed to allow the incorporation of variability in the positioning of the separating boundary between the data classes.…”

Neural Networks Based on Synthesized Training Data for the Automated Detection of Chemical Plumes in Passive Infrared Multispectral Images

“…Because IR-active chemical species have characteristic vibrational frequencies, with their absorption or emission spectral signatures superimposed on the collected ground radiance, target chemical species can be distinguished. 5 The Airborne Spectral Photometric Environmental Collection Technology (ASPECT) program operated by the U. S. Environmental Protection Agency utilizes a downwardlooking multispectral imaging instrument mounted on an aircraft to detect VOCs in the atmosphere released from ground sources. The output of this imaging system is a set of radiance intensities within the wavelength region of 8-12 µm through the utilization of eight optical bandpass filters mounted on the detector.…”

“…Because IR-active chemical species have characteristic vibrational frequencies, with their absorption or emission spectral signatures superimposed on the collected ground radiance, target chemical species can be distinguished. 5…”

Quantitative Determination of Methanol and Ethanol with Synthetic Calibration Spectra in Passive Fourier Transform Infrared Remote Sensing Measurements

“…The concept of using synthetic spectra based on an assumed radiometric model to aid multivariate calibration has been demonstrated previously. 10,11 Using synthetic spectral data enables flexibility in changing key input parameters, such as the chemical components (analytes or interferences) in the FOV, analyte concentrations, and the temperatures of the background and target cloud. When building a calibration model, similarity between the calibration data and any prediction data to which the model will be applied is important.…”