The accurate on-orbit radiometric calibration of optical sensors has become a challenge for both ensuring the consistency of space measurements and reaching the absolute accuracy required by increasingly demanding scientific requirements. Different targets are traditionally used for calibration depending on the sensor or spacecraft specificities: from on-board calibration systems to ground targets, they all take advantage of our capacity to characterize and model them. Thanks to their agility, some satellites have the capability to view extra-terrestrial targets such as the moon or stars taking advantage of the absence of atmosphere. The moon is widely used for calibration and its albedo is known through ROLO (RObotic Lunar Observatory) USGS model but its limited accuracy constrains its use to sensor drift monitoring or cross-calibration. The spectral irradiance of some stars is known with a very high accuracy, providing an absolute reference for remote sensors calibration. But the low irradiance of stars requires an instrument with a small Instantaneous Field Of View to observe them. A good knowledge of the instrument’s Modulation Transfer Function (MTF) is also necessary to perform an accurate radiometric calibration. This paper describes a method based on stars for simultaneously computing the radiometric calibration of PLEIADES 1B’ high resolution optical sensor and its MTF. The radiometric model is solved in Fourier space for point sources whose irradiance is controlled. Results are compared to the official MTF and radiometric calibration based on Pseudo Invariant Calibration Sites (PICS) and the moon. The quality of long time series of measurements is discussed as well as their accuracy.
The accurate on orbit radiometric calibration of optical sensors has become a challenge for space agencies which have developed different technics involving on-board calibration systems, ground targets or extra-terrestrial targets. The combination of different approaches and targets is recommended whenever possible and necessary to reach or demonstrate a high accuracy. Among these calibration targets, the moon is widely used through the well-known ROLO (RObotic Lunar Observatory) model developed by USGS. A great and worldwide recognized work was done to characterize the moon albedo which is very stable. However the more and more demanding needs for calibration accuracy have reached the limitations of the model. This paper deals with two mains limitations: the residual error when modelling the phase angle dependency and the absolute accuracy of the model which is no more acceptable for the on orbit calibration of radiometers. Thanks to PLEIADES high resolution satellites agility, a significant data base of moon and stars images was acquired, allowing to show the limitations of ROLO model and to characterize the errors. The phase angle residual dependency is modelled using PLEIADES 1B images acquired for different quasi-complete moon cycles with a phase angle varying by less than 1°. The absolute albedo residual error is modelled using PLEIADES 1A images taken over stars and the moon. The accurate knowledge of the stars spectral irradiance is transferred to the moon spectral albedo using the satellite as a transfer radiometer. This paper describes the data set used, the ROLO model residual errors and their modelling, the quality of the proposed correction and show some calibration results using this improved model.
The accurate on-orbit radiometric calibration of optical sensors has become a challenge for space agencies who gather their effort through international working groups such as CEOS/WGCV or GSICS with the objective to insure the consistency of space measurements and to reach an absolute accuracy compatible with more and more demanding scientific needs. Different targets are traditionally used for calibration depending on the sensor or spacecraft specificities: from on-board calibration systems to ground targets, they all take advantage of our capacity to characterize and model them. But achieving the in-flight stability of a diffuser panel is always a challenge while the calibration over ground targets is often limited by their BDRF characterization and the atmosphere variability. Thanks to their agility, some satellites have the capability to view extra-terrestrial targets such as the moon or stars. The moon is widely used for calibration and its albedo is known through ROLO (RObotic Lunar Observatory) USGS model but with a poor absolute accuracy limiting its use to sensor drift monitoring or cross-calibration. Although the spectral irradiance of some stars is known with a very high accuracy, it was not really shown that they could provide an absolute reference for remote sensors calibration. This paper shows that high resolution optical sensors can be calibrated with a high absolute accuracy using stars. The agile-body PLEIADES 1A satellite is used for this demonstration. The star based calibration principle is described and the results are provided for different stars, each one being acquired several times. These results are compared to the official calibration provided by ground targets and the main error contributors are discussed.
A new permanently instrumented radiometric calibration site for high/medium resolution imaging satellite sensors is currently under development, focussing on the visible and near infra-red parts of the spectrum. The site will become a European contribution to the Committee on Earth Observation Satellites (CEOS) initiative RadCalNet (Radiometric Calibration Network). The exact location of the permanent monitoring instrumentation will be defined following the initial site characterisation. The new ESA/CNES RadCalNet site will have a robust uncertainty budget and its data fully SI traceable through detailed characterisation and calibration by NPL of the instruments and artefacts to be used on the site. This includes a CIMEL sun photometer (the permanent instrumentation) an ASD FieldSpec spectroradiometer, Gonio Radiometric Spectrometer System (GRASS), and reference reflectance standards.
ScaRaB (SCAnner for RAdiation Budget) is the name of three radiometers whose two first flight models have been
launched in 1994 and 1997. The instruments were mounted on-board Russian satellites, METEOR and RESURS. On
October 12th, a last model has been launched from the Indian site of Sriharikota. ScaRaB is a passenger of MEGHATROPIQUES,
an Indo-French joint Satellite Mission for studying the water cycle and energy exchanges in the tropics.
The orbit is circular inclined 20deg.
ScaRaB is compatible with CERES mission. Two main spectral bands are measured by the radiometer: A short-wave
(SW) channel (0.2 – 4 μm) dedicated to solar fluxes and a Total (Tot) channel (0.2 – 200 μm) for (total) fluxes
combining the infrared earth radiance and the albedo. The earth long-wave (LW) radiance is isolated by subtracting the
SW channel to the Total channel. Thus is defined a supplemental (virtual) channel.
La Crau test site is used by CNES since 1987 for vicarious calibration of SPOT cameras. The former calibration
activities were conducted during field campaigns devoted to the characterization of the atmosphere and the site
reflectances. Since 1997, au automatic photometric station (ROSAS) was set up on the site on a 10m height pole. This
station measures at different wavelengths, the solar extinction and the sky radiances to fully characterize the optical
properties of the atmosphere. It also measures the upwelling radiance over the ground to fully characterize the surface
reflectance properties. The photometer samples the spectrum from 380nm to 1600nm with 9 narrow bands. Every non
cloudy days the photometer automatically and sequentially performs its measurements. Data are transmitted by GSM
(Global System for Mobile communications) to CNES and processed. The photometer is calibrated in situ over the sun
for irradiance and cross-band calibration, and over the Rayleigh scattering for the short wavelengths radiance
calibration. The data are processed by an operational software which calibrates the photometer, estimates the atmosphere
properties, computes the bidirectional reflectance distribution function of the site, then simulates the top of atmosphere
radiance seen by any sensor over-passing the site and calibrates it.
This paper describes the instrument, its measurement protocol and its calibration principle. Calibration results are
discussed and compared to laboratory calibration. It details the surface reflectance characterization and presents SPOT4
calibration results deduced from the estimated TOA radiance. The results are compared to the official calibration.
In partnership with the European Commission and in the frame of the Global Monitoring for Environment and Security
(GMES) program, the European Space Agency (ESA) is developing the Sentinel-2 optical imaging mission devoted to
the operational monitoring of land and coastal areas.
The Sentinel-2 mission is based on a satellites constellation deployed in polar sun-synchronous orbit. While ensuring
data continuity of former SPOT and LANDSAT multi-spectral missions, Sentinel-2 will also offer a wide range of
improvements such as a global coverage, a large field of view (290km), a high revisit capability (5 days with two
satellites), a high resolution (10m, 20m and 60m) and multi-spectral imagery (13 spectral bands). In this context, the
Centre National d'Etudes Spatiales (CNES) supports ESA to define the system image products and to prototype the
relevant image processing techniques.
First, this paper presents the Sentinel-2 system and the image products that will be delivered: starting from raw
decompressed images up to accurate ortho-images in Top of Atmosphere reflectances. The stringent image quality
requirements are also described, in particular the very accurate target geo-location.
Then, the prototyped image processing techniques will be addressed. Both radiometric and geometric processing will be
described with a special focus on the automatic enhancement of the geometric physical model involving a global set of
reference data.
Finally, the very promising results obtained by the prototype will be presented and discussed. The radiometric and
geometric performances will be provided as well as the associated computing time estimation on the target platform.
In the framework of the Global Monitoring for Environment and Security (GMES) programme, the European Space
Agency (ESA) in partnership with the European Commission (EC) is developing the SENTINEL-2 optical imaging
mission devoted to the operational monitoring of land and coastal areas. The Sentinel-2 mission is based on a twin
satellites configuration deployed in polar sun-synchronous orbit and is designed to offer a unique combination of
systematic global coverage with a wide field of view (290km), a high revisit (5 days at equator with two satellites), a
high spatial resolution (10m, 20m and 60 m) and multi-spectral imagery (13 bands in the visible and the short wave
infrared spectrum). SENTINEL-2 will ensure data continuity of SPOT and LANDSAT multispectral sensors while
accounting for future service evolution.
This paper presents the main geometric and radiometric image quality requirements for the mission. The strong multi-spectral
and multi-temporal registration requirements constrain the stability of the platform and the ground processing
which will automatically refine the geometric physical model through correlation technics. The geolocation of the
images will take benefits from a worldwide reference data set made of SENTINEL-2 data strips geolocated through a
global space-triangulation. These processing are detailed through the description of the level 1C production which will
provide users with ortho-images of Top of Atmosphere reflectances. The huge amount of data (1.4 Tbits per orbit) is
also a challenge for the ground processing which will produce at level 1C all the acquired data.
Finally we discuss the different geometric (line of sight, focal plane cartography, ...) and radiometric (relative and
absolute camera sensitivity) in-flight calibration methods that will take advantage of the on-board sun diffuser and
ground targets to answer the severe mission requirements.
VENµS is a demonstration mission developed in cooperation between Israël (ISA) and France (CNES). VENµS
scientific mission unique feature is to acquire high resolution (5.3m) multi-spectral images (12 bands in the visible and
NIR spectrum) continuously every second day with constant viewing angles. At least 50 sites of interest all around the
world will be viewed. It aims at demonstrating the relevance of such observation capabilities in the framework of the
European Global Monitoring for Environment and Security Program (GMES). The satellite also flies a technological
mission that aims at qualifying an Israeli electric propulsion technology (IHET) and demonstrating its mission
enhancement capabilities. The satellite will be launched in January 2010. The imaging scientific mission will last 2.5
years with the satellite at 720 km. Next, the technological mission will bring the satellite at 410 km. The scientific
mission will then go on for one year with an improved resolution (3m).
This paper presents the main geometric and radiometric image quality requirements for the scientific mission. The strong
multi-spectral (2m) and multi-temporal (3m) registration requirements constrain the stability of the platform and the
ground processing which will refine the geometric physical model using an image matching method based on
correlation. The location of the images will take benefits from the capacity of the system to produce Digital Elevation
Models at a low 'Base to Elevation' ratio (0.026). These processings are detailed through the description of the level 1
production which will provide users with ortho-images of Top of Atmosphere reflectances.
Finally we propose different radiometric (relative and absolute camera sensitivity,...) and geometric (line of sight, focal
plane cartography,...) in-flight calibration methods to answer the severe mission requirements.
This paper reviews the calibration techniques used for 19 years of SPOT satellite exploitation:
- from the decision to fly a calibration system (involving a lamp and a sun sensor) on SPOT1 to its suppression on SPOT5 and to the emphasis placed on the development of calibration methods over natural targets like oceans or deserts. Oceans provide low reflectance targets for short wavelengths calibration over the atmospheric molecular scattering, while deserts (warm over North Africa and Middle East, cold over Antarctica) provide stable references for sensors cross-calibration and temporal monitoring.
- from vicarious calibration campaigns performed by scientific teams over test sites like White Sands (USA) or La Crau (France) to an autonomous ground based station. An automatic radiometer continuously characterizes the reflectance and the atmosphere of the French test site and thanks to its original calibration procedure provides the top of atmosphere radiance needed for in-flight calibration.
The results provided by these different methods are discussed. We show that the on-board calibration unit used to monitor with time SPOT1, 2, 3 and 4 cameras sensitivity is loosing sensitivity, justifying the overall calibration update that was proposed to users in October 2004 for all SPOT since their beginning of life.
SPOT5, the fifth satellite of the SPOT remote sensing satellite family was successfully launched on the 4th of May 2002. SPOT5 is designed to ensure continuity of data acquisition and space image services but also to provide users with advanced products. It flies two identical cameras named HRG (High Resolution Geometry) providing a 2.5 m and a 5 m resolution in a panchromatic mode and a 10 m resolution in a multi-spectral mode, still keeping a 60-km ground field. Stereo application is part two of the SPOT5 mission; the satellite flies a specific High Resolution Stereo instrument (HRS) made up of two telescopes allowing a 20° fore view and a 20° aft view over a 120-km swath, sampling the landscape every 5m. VEGETATION2, a wide field of view imaging radiometer complements the mission thanks to its daily coverage of the earth. The paper presents the mission, the commissioning phase that followed the satellite launch, the assessment of the image quality and the first calibration results.
The SPOT5 remote sensing satellite was launched in May 2002. It provides SPOT service continuity above and beyond SPOT4 operation but the SPOT5 system also significantly improves the SPOT service with the new characteristics of its two HRG (High Resolution Geometry) cameras and its HRS (High Resolution Stereo) camera. SPOT5's first two months of life in orbit were dedicated to instrument calibration and the assessment of image quality performances. During this period, the CNES team used specific target programming to compute image correction parameters and estimate the performance of the image processing chain, at system level. This paper focuses on the relative radiometric performances of the different spectral bands for the three instruments, deduced from in-flight measurements. For each spectral band, a radiometric model gives the ratio between detector response and input radiance. This model takes the architecture of the onboard image chain into account. Calibration provides the normalisation parameters (dark currents and relative inter-detector sensitivities) used to correct the images. The quality of the corrected images is quantified through several signal-to-noise ratio measurements based on different techniques. These methods are presented and their accuracy is discussed. Finally, a comparison is given between flight measurements and ground measurements.
The MTF (Modulation Transfer Function) is a means of characterizing the spatial resolution of the instruments. So, the MTFs of HRG and HRS cameras are parts of image quality parameters assessed during the in-flight commissioning phase. Vibrations during the launch and transition from air to vacuum may defocus the HRG cameras and degrade their MTF. Therefore, SPOT5 HRG cameras are refocused before measuring their MTF. The paper first describes the HRG focusing procedure that uses both cameras viewing the same landscape: the focus of one camera is changed while the other is fixed and used as a reference. Results are given for each camera in terms of best focus and focus variation in the field of view. These results are compared to those provided by an autotest system, on-board each HRG camera, that images a high frequency periodic pattern while the focus is changed. Then, MTF measurements are presented. The MTF of HRG cameras is measured by imaging a spotlight that aimed at the satellite; the results are compared with pre-flight measurements. Besides, the MTF of HRS cameras is assessed by imaging landscapes with edge patterns; the main objective is to compare the two HRS cameras.
In the characterization of a space-borne wide field-of-view sensor, like Végétation, the multi-angular calibration is strongly complementary to the absolute calibration. It is defined as the process of estimating the sensitivity variations at different points of the Végétation wide field-of-view. This effect has to be integrated in the data processing. Pre-flight measurements were performed before launch, but because of heavy irradiations and aging of the different part of the sensor, it is necessary after launch to check and/or adjust the multi-angular calibration coefficients, gp. For this, the gp coefficients were split into three terms which required different methods: i/ first, the low-frequency term (gpLF) which refer to variation of the optic transmission which slightly decreases when viewing angle increases. The gpLF were verified using acquisitions over 20 desert sites for which TOA reflectances are accurately characterized (from ground measurements and POLDER/ADEOS-1 measurements). No in-flight variation of the gpLF were detected. ii/ second, the high-frequency term (gpHF) which refer to variation of the sensitivity of the elementary detectors. The gpHF were verified statistically using acquisitions over the Antarctica site and were accurately checked for the 4 spectral bands. ii/ third, the medium-frequency term (gpMF) which refer to various kinds of variation (optics, detectors...). The gpMF were verified during the 9pJpWDWLRQ like using the on-board calibration device (lamp profiles) and some small variations were identified (< 0.5% for B0, B2, B3 and ~1% for MIR). This aspect is still under investigation using acquisitions over Antarctica.
The Rayleigh scattering over a clear ocean is a target which radiance is very well modeled and which enables to calibrate the short wavelengths of remote sensing instruments. But the quality of the calibration strongly depends on the evaluation of the other contributors to the observed Top Of Atmosphere radiance i. e. aerosol scattering and reflection over the sea surface (water color, foam, glint...). However these contributors can be reduced by appropriate viewing conditions. This technique is used to calibrate B1 (051-0.59 µm) and B2 (0.61-0.68µm) channels of HRVIR camera, and B0 (0.4-0.5µm) and B2 channels of VEGETATION camera both of which are aboard SPOT4. This article presents the calibration results obtained during the satellite two years in orbit. The results are compared to: - pre-flight results (integrating sphere) - in-flight results. The in-flight results are provided by: - on board calibration system (lamp and sun sensor) - vicarious calibration over test sites (White Sands, La Crau) - calibration over stable deserts - calibration over the sun glint The analysis of the sensitivity of the calibration to the different parameters used to model the TOA radiance shows the accuracy of such a technique.
A single data set of spatially extensive hyperspectral imagery is used to carry out vicarious calibrations for multiple Earth observation sensors. Results are presented based on a data acquisition campaign at the newell County rangeland test site in Alberta in October 1998, which included ground-based measurements, satellite imagery, and airborne casi hyperspectral data. This paper present new calibration monitoring obtained for NOAA-14 AVHRR, OrbView-2 SeaWiFS, SPOT-4 VGT, Landsat-5 TM, and SPOT-2 HRV.
The SPOT4 remote sensing satellite was successfully launched at the end of March 1998. It was designed first of all to guarantee continuity of SPOT services beyond the year 2000 but also to improve the mission. Its two cameras are now called HRVIR since a short-wave infrared (SWIR) spectral band has been added. Like their predecessor HRV cameras, they provide 20-meter multispectral and 10-meter monospectral images with a 60 km swath for nadir viewing. SPOT4's first two months of life in orbit were dedicated to the evaluation of its image quality performances. During this period of time, the CNES team used specific target programming in order to compute image correction parameters and estimate the performance, at system level, of the image processing chain. After a description of SPOT4 system requirements and new features of the HRVIR cameras, this paper focuses on the performance deduced from in-flight measurements, methods used and their accuracy: MTF measurements, refocusing, absolute calibration, signal-to-noise Ratio, location, focal plane cartography, dynamic disturbances.
SPOT4, the fourth satellite of the SPOT family remote sensing satellites, was launched on the 20th of March 1998. During the first months, we calibrate the two identical on-board cameras named HRVIR (because of the added Mid Infra-Red channel) and VEGETATION, a wide field of view radiometer providing 1.15 kilometers resolution measurements in the same designed channels as HRVIR (B2, B3 and MIR), and we evaluate the quality of the images. Radiometric calibration results are presented in this paper. Different methods are applied based on the experience gained with SPOT1, 2, 3 and POLDER: (1) pre- launch measurements, (2) on-board calibration system, (3) vicarious calibration over test sites, (4) inter-SPOT calibration over desert areas, (5) calibration over the molecular scattering, (6) inter-cameras calibration between HRVIR1 and HRVIR2, (7) inter-cameras calibration between HRVIR and VEGETATION. The accuracy of each calibration procedure is estimated. The measurements are combined in a model that minimizes errors and provides the camera sensitivity as a function of time.
Olivier Hagolle, Philippe Goloub, Pierre-Yves Deschamps, T. Bailleul, Jean-Marie Nicolas, Yves Fouquart, Aime Meygret, Jean Luc Deuze, Maurice Herman, Frederic Parol, Francois-Marie Breon
POLDER is a CNES instrument on-board ADEOS polar orbiting satellite, which was successfully launched in August 1996. In November 1996, POLDER entered its nominal acquisition phase and functioned perfectly until ADEOS early end of service in June 1997. POLDER is a multispectral imaging radiometer/polarimeter designed to collect global and repetitive observations of the solar radiation reflected by the Earth/atmosphere system, with a wide field of view (2400 km) and a moderate geometric resolution (6 km). The instrument concept is based on telecentric optics, on a rotating wheel carrying 15 spectral filters and polarizers, and on a bidimensional CCD detector array. In addition to the classical measurement and mapping characteristics of a narrow-band imaging radiometer, POLDER has a unique ability to measure polarized reflectances using three polarizers (for three of its eight spectral bands, 443 to 910 nm), and to observe target reflectances from 13 different viewing directions during a single satellite pass. One of POLDER original features is that its in-flight radiometrical calibration does not rely on any on-board device. Many calibration methods using well-characterized calibration targets have been developed to achieve a very high calibration accuracy. This paper presents the various methods involved in the absolute in-flight calibration plan and the results obtained during the calibration phase of the instrument: absolute calibration over molecular scattering, inter-band calibration over sunglint and clouds, inter-calibration with OCTS, water vapor channels calibration over sunglint using meteorological analysis. A brief description of the algorithm and of the performances of each method is given.
Vicarious calibration generally requires field activities in order to characterize surface reflectances and atmosphere to unable use the prediction of the radiance at the satellite level. To limit human presence on the field, an automatic ground-based station was defined as well as the required protocol to achieve satellite vicarious calibration. The solar irradiance measurements are self calibrated using the Langley technique. The instrument was designed so that, firstly, the same gun measures both the solar irradiance and the radiance (sky or ground) and, secondly, that the field of view is constant over the spectral range. These two conditions offer an intercalibration opportunity between radiance and irradiance as well as the field of view is well defined. Experimental determination of the field of view is possible in UV region based on the Rayleigh scattering. We, then, describe how to derive the TOA signal from measurements. Two approaches have been developed according the energetic characteristics we want to estimate (reflectance or radiance). Preliminary results of a field campaign in June 1997 are reported.
Since mid-87, the histograms of all the SPOT cloud-free images processed in Toulouse (France), Kiruna (Sweden) and by SICORP (USA) have been stored in a data base. Thanks to the continuous effort made to calibrate the SPOT HRVs, the observed reflectances for the four SPOT spectral bands can be estimated. Reflectances can be visualized on world maps for a given period and are mainly used to tune the viewing gains. This calibration procedure is now operational for SPOT2 and SPOT3: the new programming center automatically adjusts the on-board electronic gains according to the landscape to optimize the image dynamics while avoiding saturation. This paper presents the basic design of the base, from the reception of histograms to the creation of gain files. Reflectance maps show the data base contents in terms of world coverage. The accuracy of the computed reflectances is discussed and the problem of geographic areas with no data in the base in considered. We also briefly describe other SPOT Histogram Data Base applications such as sizing an optical sensor, and searching for uniform areas for relative calibration or stable areas for absolute temporal calibration. Finally we discuss future enhancements such as using data from other sensors like VEGETATION, which will fly on SPOT4 and would provide information for the mid- infrared band to optimize SPOT4 HRVIR image acquisition.
A successful in-flight refocusing experiment based on image processing is described. Each of the two SPOT1 HRV cameras was refocused with respect to the other by analyzing the image spectrum taken simultaneously by both cameras. The experiment was carried out during the autumn of 1994 and its results are also presented: (1) the estimated optimal position for the rear corrective lens of the camera, (2) MTF improvement and its relative quality in the field of view in terms of homogeneity and astigmatism, (3) the validation of a theoretical geometric defocusing model that gives the changes in MTF as a function of corrective lens position. We conclude with the high accuracy of our method (2%) and its sensitivity to the spectral content of the viewed landscapes.
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.