In the field of target detection, the simulation of the camera FOV (field of view) background is a significant issue. The presence of heterogeneous clouds might have a strong impact on a target detection algorithm. In order to address this issue, we present here the construction of the CERAMIC package (Cloudy Environment for RAdiance and MIcrophysics Computation) that combines cloud microphysical computation and 3D radiance computation to produce a 3D atmospheric infrared radiance in attendance of clouds.
The input of CERAMIC starts with an observer with a spatial position and a defined FOV (by the mean of a zenithal angle and an azimuthal angle). We introduce a 3D cloud generator provided by the French LaMP for statistical and simplified physics. The cloud generator is implemented with atmospheric profiles including heterogeneity factor for 3D fluctuations. CERAMIC also includes a cloud database from the French CNRM for a physical approach. We present here some statistics developed about the spatial and time evolution of the clouds. Molecular optical properties are provided by the model MATISSE (Modélisation Avancée de la Terre pour l’Imagerie et la Simulation des Scènes et de leur Environnement).
The 3D radiance is computed with the model LUCI (for LUminance de CIrrus). It takes into account 3D microphysics with a resolution of 5 cm-1 over a SWIR bandwidth. In order to have a fast computation time, most of the radiance contributors are calculated with analytical expressions. The multiple scattering phenomena are more difficult to model. Here a discrete ordinate method with correlated-K precision to compute the average radiance is used. We add a 3D fluctuations model (based on a behavioral model) taking into account microphysics variations. In fine, the following parameters are calculated: transmission, thermal radiance, single scattering radiance, radiance observed through the cloud and multiple scattering radiance.
Spatial images are produced, with a dimension of 10 km x 10 km and a resolution of 0.1 km with each contribution of the radiance separated. We present here the first results with examples of a typical scenarii. A 1D comparison in results is made with the use of the MATISSE model by separating each radiance calculated, in order to validate outputs. The code performance in 3D is shown by comparing LUCI to SHDOM model, referency code which uses the Spherical Harmonic Discrete Ordinate Method for 3D Atmospheric Radiative Transfer model. The results obtained by the different codes present a strong agreement and the sources of small differences are considered. An important gain in time is observed for LUCI versus SHDOM. We finally conclude on various scenarios for case analysis.
The availability of very high spatial resolution sensors has for the past few years allowed a precise description of atmospheric scenes for remote sensing and surveillance applications. Clouds presence in the field of view is one of the key factors limiting the performances of these sensors for target detections. However, in order to develop such detection algorithms for images with a fine spatial resolution, a fast 3D radiative transfer tool dedicated to scene generation is necessary to obtain large number of realistic cloud scenes. Three-dimensional effects become more important when going to higher model resolution. For that purpose, fast solutions are needed since three-dimensional radiative transfer solvers are computationally far too expensive. Two different strategies are presented in this paper. On the one hand, an optimization of the explicit method Spherical Harmonic Discrete Ordinate Method (SHDOM) developed by Evans, K. F. (1998), associated with a fast image rendering solution. On the other hand, a fast approximation of 3D radiative transfer.
In this paper, we present the process of validating the atmospheric modelling software MATISSE (Advanced Modelling
of the Earth for the Imaging and the Simulation of the Scenes and their Environment, developed by ONERA) by
comparing simulation results of terrestrial and atmospheric background to MODIS satellite images. Analyses have been
carried out for wavebands in the visible (VIS) as well as the longwave infrared (LWIR) spectrum.
Generally available satellite images, e. g. from the MODIS sensor, provide data in spectral bands, which are suitable for
remote sensing applications and earth surface observations. However, for some applications different bands as well as
specific cloud formations for a certain region may be of interest, thus making the simulation of background data
essential. Therefore, the software MATISSE ("Advanced Modeling of the Earth for Environment and Scenes
Simulation") proved to be the appropriate tool. MATISSE is an infrared background scene generator developed by
ONERA for computing natural background spectral radiance images including atmosphere, sea, land and high and low
altitude clouds. In order to validate the model, comparisons with MODIS satellite data have been carried out using
images in available spectral bands. The investigations comprised selected surface structures like sea, desert, lowland
(dry) and highlands (humid). In general, the results on radiance images show a good correlation between MODIS image
and the MATISSE-simulation.
This paper focuses on comparing results between simulated MATISSE radiance images and the MODIS observations.
Based on this, possible sources of error and the limits of the model are discussed.
The validation of the sea surface infrared optical properties multiresolution model developed at ONERA is investigated
by comparison with measurements. The images were obtained during the MIRAMER campaign that took place in May
2008 in the Mediterranean Sea. The sea radiance model and optical properties are expressed and the experimental setup
of the campaign is briefly presented. We focus on solar glint measurements collected the 22 th of May at 5 h 59mn 50s
in the MWIR bandwidth on-board the R/V ATALANTE at grazing observational angle. A sensitivity analysis of glitter
radiance on atmospheric and aerosol profiles in the vicinity of the measured contextual parameters is presented. Modelled
and measured images are compared and results are delved further by comparisons of histograms, averaged vertical and
horizontal profiles. Errors are under those potentially due to calibration. Finally, a sensitivity analysis upon uncertainties
on contextual parameters involved in sea radiance computation is made.
MATISSE (Advanced Modeling of the Earth for Environment and Scenes Simulation) is an infrared background scene
generator developed for computing natural background spectral radiance images. The code also provides atmospheric
radiatives quantities along lines of sight. Spectral bandwidth ranges from 0.4 to 14 μm. Natural backgrounds include
atmosphere, sea, land and high and low altitude clouds. The new version MATISSE-v2.0, released this year, has been
designed to treat spatial multi resolution in the generated images in order to be able to reach metric spatial variability in
pixels footprints. Moreover, MATISSE-v2.0 includes a new sea surface radiance model (water waves and surface optical
properties) which depends on wind speed, wind direction and fetch value. Preliminary validations using radiometric
measurements have been conducted concerning sea radiances and give promising results. In order to go further in the
validation process of MATISSE-v2.0, comparisons with MODIS satellite images have been led. The results of
comparing the simulated MATISSE images radiances with the MODIS observations show that the code is performing
well. This paper gives a description of MATISSE-v2.0 new functionalities and focus on first results on comparison
between MATISSE/MODIS images radiances.
MATISSE which acronym means Advanced Modeling of the Earth for Environment and Scenes Simulation is an
infrared background scene generator developed by Onera since the mid 1990'. MATISSE main goal is to compute
radiance images of natural backgrounds and radiative quantities such as local illumination, spectral transmission, and
spectral radiance along lines of sight.
The new version MATISSE-v2.0 has been completed during the first quarter of 2010 and the public version is going to
be released in few weeks. This latest version uses a multi resolution spatial scheme in order to treat the natural
backgrounds with spatial footprint from kilometre sizes (satellite viewing) down to metric sizes. Up to now, this spatial
scheme has been used in order to generate infrared images of sea surface. The new sea surface model (water waves and
surface optical properties) has been partially validated by using a specific Mediterranean campaign. MATISSE-v2.0 is
also accompanied with a new set of GUI (graphical user interface) in order to help the user in defining its computational
case. The code is also designed in order to be interfaced with other applications.
Our presentation will be devoted to a description of MATISSE-v2.0 new features, with examples of sea surface scenes
exemplifying the new code functionalities.
MATISSE is a background scene generator developed for the computation of natural background spectral radiance
images and useful atmospheric radiatives quantities (radiance and transmission along a line of sight, local illumination,
solar irradiance ...). The spectral bandwidth ranges from 0.4 to 14 μm. Natural backgrounds include atmosphere (taking
into account spatial variability), low and high altitude clouds, sea and land. The current version MATISSE-v1.5 can be
run on SUN and IBM workstations as well as on PC under Windows and Linux environment. An IHM developed under
Java environment is also implemented.
MATISSE-v2.0 recovers all the MATISSE-v1.5 functionalities, and includes a new sea surface radiance model
depending on wind speed, wind direction and the fetch value. The release of this new version in planned for April 2009.
This paper gives a description of MATISSE-v1.5 and MATISSE-v2.0 and shows preliminary comparison results
between generated images and measured images during the MIRAMER campaign, which hold in May 2008 in the
Mediterranean Sea.
The SE-WORKBENCH workshop, also called CHORALE (French acceptation for "simulated Optronic Acoustic Radar
battlefield") is used by the French DGA (MoD) and several other Defense organizations and companies all around the
World to perform multi-sensors simulations. CHORALE enables the user to create virtual and realistic multi spectral 3D
scenes that may contain several types of target, and then generate the physical signal received by a sensor, typically an
IR sensor.
The SE-WORKBENCH can be used either as a collection of software modules through dedicated GUIs or as an API
made of a large number of specialized toolkits.
The SE-WORKBENCH is made of several functional block: one for geometrically and physically modeling the terrain
and the targets, one for building the simulation scenario and one for rendering the synthetic environment, both in real and
non real time.
Among the modules that the modeling block is composed of, SE-ATMOSPHERE is used to simulate the atmospheric
conditions of a Synthetic Environment and then to integrate the impact of these conditions on a scene. This software
product generates an exploitable physical atmosphere by the SE WORKBENCH tools generating spectral images. It
relies on several external radiative transfer models such as MODTRAN V4.2 in the current version.
MATISSE [4,5] is a background scene generator developed for the computation of natural background spectral radiance
images and useful atmospheric radiative quantities (radiance and transmission along a line of sight, local illumination,
solar irradiance ...). Backgrounds include atmosphere, low and high altitude clouds, sea and land. A particular
characteristic of the code is its ability to take into account atmospheric spatial variability (temperatures, mixing ratio, etc)
along each line of sight. An Application Programming Interface (API) is included to facilitate its use in conjunction with
external codes.
MATISSE is currently considered as a new external radiative transfer model to be integrated in SE-ATMOSPHERE as a
complement to MODTRAN. Compared to the latter which is used as a whole MATISSE can be used step by step and
modularly as an API: this can avoid to pre compute large atmospheric parameters tables as it is done currently with
MODTRAN. The use of MATISSE will also enable a real coupling between the ray tracing process of the SEWORKBENCH
and the radiative transfer model of MATISSE. This will lead to the improvement of the link between a
general atmospheric model and a specific 3D terrain.
The paper will demonstrate the advantages for the SE WORKEBNCH of using MATISSE as a new atmospheric code,
but also for computing the radiative properties of the sea surface.
Interaction of light with atmospheric components is a major problem studied in the field of remote sensing and
atmospheric propagation. Generally, mean atmospheric profiles are used to compute radiative transfer. But here, the
sensitivity of radiances and transmittances to the climatic conditions is tested. The goal of this study is to evaluate
and quantify atmospheric components variations that influence radiances and transmittances for airborne limb
viewing observations. In particular, radiance variations are analyzed according to the variations of temperature and
of different atmospheric molecules. This radiance variability is related to the water content and temperature
modifications.
This paper concerns the presentation of the MATISSE-v1.4 code whose main functionality is computation of spectral or integrated natural background radiance images. The spectral bandwidth ranges from 765 to 3300 cm-1 (3 to 13 μm) with a 5 cm-1 resolution. Natural backgrounds include the atmosphere, low and high altitude clouds, sea and land. The most particular functionality of the code is to take into account atmospheric spatial variability quantities (temperatures, mixing ratio, etc) along each line of sight of the image. In addition to image generation capacity, the code computes atmospheric radiance and transmission along a line of sight with the same spectral characteristics as in imaging mode. In this case, atmospheric refraction effects and radiation from high or low altitude clouds can be taken into account. A high spectral resolution mode is also available to propagate radiation from a high temperature medium in the same atmospheric state as that used for the image generation. Moreover, an Application Programming Interface (API) is included to facilitate its use in conjunction with external codes. In comparison with the previous version, the main improvement of MATISSE-v1.4 concerns the line of sight mode, the possibility to use a user atmospheric profile and computations in the Maritime Boundary Layer. This paper describes the range of functionalities of MATISSE-v1.4 as well as future developments.
This paper presents the MATISSE-v1.4 code whose main functionality is to compute spectral or integrated natural background radiance images. The spectral bandwidth extends from 765 to 3300 cm-1 (3 to 13 μm) with a 5 cm-1 resolution. Natural backgrounds include the atmosphere, low and high altitude clouds, sea and land. The most particular functionality of the code is to take into account atmospheric spatial variability quantities (temperatures, mixing ratio, etc) along each line of sight of the image. In addition to image generation capacity, the code computes atmospheric radiance and transmission along a line of sight with the same spectral characteristics as in imaging mode. In this case atmospheric refraction effects and radiation from high or low altitude clouds can be taken into account. A high spectral resolution mode is also available to propagate radiation from a high temperature medium in the same atmospheric state as that used for the image generation. Finally, an Application Programming Interface (API) is included to facilitate its use in conjunction with external codes. This paper describes the range of functionalities of MATISSE-v1.4 whose release is planned for April 2006. Future developments are also presented.
In this paper we present MATISSE 1.1 a new background scene generator, whose goal is to compute spectral or integrated radiance images of natural background, as well as the transmission of a hot gas signature.
The spectral bandwidth for this version of the code is from 750 to 3300 cm-1 (3 to 13 μm) with a 5 cm-1 resolution. Gaseous absorption is computed by a Correlated K model. The spatial variability of atmospheric quantities (temperatures and mixing ratios, among others) is taken into account, using variable profiles along the line of sight.
Natural backgrounds include the atmospheric background, low altitude clouds and the Earth ground. The radiation models used are designed for observation at low spatial resolution of clouds and soils, so a texture model was developed to increase the high spatial resolution rendering in the metric range.
Intermediate outputs of the code deliver radiance and transmission restricted to a single line of sight, in which case atmospheric refraction effects are taken into account. Along this line of sight the transmission can also be computed using a line-by-line model, which is useful to propagate the radiation emitted by a hot gas source (fires, aircraft or missile plume).
MATISSE 1.1 was released in June 2002, so this paper is devoted to a presentation of the first results obtained with the code and some validation tests.
MATISSE is a new atmospheric radiative transfer code currently under development at Onera. Its purpose is to compute background radiance images by taking into account atmospheric, cloud and ground radiation and the variability of atmospheric properties. Propagation is calculated using a Correlated K model (CK) developed at Onera. The spectral range is between 3 to 13 micrometers with a resolution of 5 cm-1. Weather forecast outputs and aerosol climatology are used as inputs to account for spatial variability of atmospheric properties in radiance computations. Partial stratocumulus cloud cover can be generated and the radiation computations use Independent Pixel Approximation (IPA) and Bidirectional Reflectivity Distribution Functions (BRDF). Ground emission and reflectance are computed from spectral emissivities, BRDF and a simple thermal model for the local ground temperature. Databases include a Digital Terrain Elevation (DTED) and a land use database with 30' spatial resolution. Texture models are used to add realistic ground and cloud clutter down to 10 meter resolution. A line-by-line model is included to compute the spectral intensity propagated from high temperature exhaust plumes. Refraction effects are computed, but only along one single line of sight.
Airborne measurement of stratus is being carried out by ONERA in order to validate NUALUM cloud simulation. The cloud top is measured by a circular variable filter cryogenic spectrometer SICAP. Two observation zenith angles are tested and the azimuth angle is variable. In situ liquid water content (LWC) measurement in performed by a Johnson Williams probe and compared with meteorological sounding. The liquid water content has been valuated inside a layer from 200m to 600m altitude. Mean LWC is equal to 0.3g/m3. Those values are typical of a stratus. The NUALUM cloud radiative transfer code has been developed at ONERA. The optical properties are computed by MIE theory. NUALUM includes the DISORT code to compute the multiple scattering in the cloud, by the mean of the discrete ordinates method. Spectral measurements show great variations of the radiation according to the azimuth angle when the wavelength is less than 4 micrometers . In this range of wavelength, radiation provides from solar reflection which is very sensitive to the scattering angle. Above 4 micrometers , thermal radiance increases. A good correlation is observed between the shape of the phase function of cloud particles and the radiance variation with scattering angle. NUALUM is in agreement with the stratus spectral measurements. During the experiment, sea spectral measurements have also been carried out. Solar reflection on the sea surface of the sea is specular and only occurs for low scattering angles.
A new model of radiative transfer in a cloudy atmosphere NUALUM has been developed at ONERA. NUALUM uses
the Discrete Ordinates Method. This code includes multiple scattering in clouds and takes into account the azimuthal
distribution of radiation. NUALUM allows geometry and microphysics of the cloud to be varied (particle size, concentration,
top and bottom altitude).
The parameterization of a cumulus cloud with NUALUM is compared with airborne measurements performed by
ONERA and CELAR: circularvariable filter cryogenic spectrometer SICAP (1.5-5.5 j.tm) and airborne infrared camera CIRAP
(3-5 pm and 8-12 tm) aim at the same point of the cloud top. The observation azimuthal angle is variable.
SICAP spectral measurements show great variations of the radiation with the azimuthal angle which are in agreement
with NUALUM simulations. The correlation between the infrared radiation images CIRAP and mean calculated radiation is
satisfactory.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.