Sidonie Lefebvre

Aggregation process of optical properties and temperature over heterogeneous surfaces in infrared domain

We propose a modeling of the aggregation processes of optical properties and temperature over the heterogeneous landscape in the infrared domain (3-14 microm). The main objectives of the modeling are to understand how these parameters aggregate and to study their links at different spatial scales. As the landscape is described at each scale by its radiative parameters, general equations linking the radiative parameters at a given high spatial scale to those at a rough scale are proposed. Then these equations are applied to several synthetic landscapes. An analysis based on a design of experiments is conducted to point out the influence of each of the input factors. The results show the importance of the intrinsic parameters (reflectance, emissivity, and surface temperature) of each surface element and also the directional and spectral behaviors of the aggregated parameters.

Relevance of an inverse problem approach to overcome cut-off wavenumbers disparities in infrared stationary Fourier transform spectrometers.

One of the major limitations to the use of infrared focal plane arrays (IRFPAs) in stationary Fourier transform spectrometers (FTSs) comes from the spatial inhomogeneities of the pixel responses, where the inhomogeneities of the cut-off wavenumbers of the pixels can prevail. The hypothesis commonly assumed for FTSs that all the pixels are equivalent is thus inaccurate and results in a degradation of the estimated spectrum, even far from the cut-off wavenumbers. However, if the individual spectral responses of the pixels are measured beforehand, this a priori information can be used in the inversion process to produce reliable spectra. Thus, spatial inhomogeneities are not an obstacle for the use of infrared stationary FTS. This result is illustrated in this paper by numerical simulations, based on a realistic description of an IRFPA.

Modeling of HgCdTe focal plane array spectral inhomogeneities

Infrared focal plane arrays (IRFPA) are widely used to perform high quality measurements such as spectrum acquisition at high rate, ballistic missile defense, gas detection, and hyperspectral imaging. For these applications, the fixed pattern noise represents one of the major limiting factors of the array performance. This sensor imperfection refers to the nonuniformity between pixels, and is partially caused by disparities of the cut-off wavenumbers. In this work, we focus particularly on mercury cadmium telluride (HgCdTe), which is the most important material of IR cooled detector applications. Among the many advantages of this ternary alloy is the tunability of the bandgap energy with Cadmium composition, as well as the high quantum efficiency. In order to predict and understand spectral inhomogeneities of HgCdTe-based IRFPA, we propose a modeling approach based on the description of optical phenomena inside the pixels. The model considers the p-n junctions as a unique absorbent bulk layer, and derives the sensitivity of the global structure to both Cadmium composition and HgCdTe layer thickness. For this purpose, HgCdTe optical and material properties were necessary to be known at low temperature (80K), in our operating conditions. We therefore achieved the calculation of the real part of the refractive index using subtracti

Hyperspectral modeling of an infrared focal plane array

Infrared Focal Plane Arrays (FPA) are increasingly used to measure multi- or hyperspectral images. Therefore, it is crucial to control and modelize their spectral response. The purpose of this paper is to propose a modeling approach, adjustable by experimental data, and applicable to the main cooled detector technologies. A physical model is presented, taking into account various optogeometrical properties of the detector, such as disparities of the pixels cut-off wavelengths. It describes the optical absorption phenomenon inside the pixel, by considering it as a stack of optical bulk layers. Then, an analytical model is proposed, based on the interference phenomenon occurring into the structure. This model considers only the three major waves interfering. It represents a good approximation of the physical model and a complementary understanding of the optical process inside the structure. This approach is applied to classical cooled FPAs as well as to specific instruments such as Microspoc (MICRO SPectrometer On Chip), a concept of miniaturized infrared Fourier transform spectrometer, integrated on a classical Mercury-Cadmium-Telluride FPA, and cooled by a cryostat.

Analysis of propellant combustion with real-time multispectral infrared camera

We present a compact real-time multispectral camera operating in the mid-infrared wavelength range. Multispectral images of a scene with two differently spectrally signed objects and of a burning solid propellant will be shown. Ability of real-time acquisition will thus be demonstrated and spectra of objects will be retrieved thanks to inversion algorithm applied on multispectral images.

Infrared focal plane array with a built-in stationary Fourier-transform spectrometer (MICROSPOC): physical limitations and numerical solutions

Microspoc is a compact Fourier transform spectrometer, with the interferometer integrated on the focal plane array. This paper discusses the way to overcome the limitations due to parasitic interferences inside the active layer of the photodetectors.

Towards a handheld cryogenic FTIR spectrometer

A new concept of Fourier-transform interferometer integrated in the focal plane array has been developed. Properties of this element, compact optical design and experimental results obtained with a prototype will be detailed.

Linear regression models and neural networks for the fast emulation of a molecular absorption code

The background scene generator MATISSE, whose main functionality is to generate natural background radiance images, makes use of the so-called Correlated K (CK) model. It necessitates either loading or computing thousands of CK coefficients for each atmospheric profile. When the CK coefficients cannot be loaded, the computation time becomes prohibitive. The idea developed in this paper is to substitute fast approximate models for the exact CK generator; using the latter, a representative set of numerical examples is built and used to train linear or nonlinear regression models. The resulting models enable an accurate CK coefficient computation for all the profiles of an image in a reasonable time.