Search results for "gap filling"

showing 6 items of 6 documents

Multispectral high resolution sensor fusion for smoothing and gap-filling in the cloud

2020

Remote sensing optical sensors onboard operational satellites cannot have high spectral, spatial and temporal resolutions simultaneously. In addition, clouds and aerosols can adversely affect the signal contaminating the land surface observations. We present a HIghly Scalable Temporal Adaptive Reflectance Fusion Model (HISTARFM) algorithm to combine multispectral images of different sensors to reduce noise and produce monthly gap free high resolution (30 m) observations over land. Our approach uses images from the Landsat (30 m spatial resolution and 16 day revisit cycle) and the MODIS missions, both from Terra and Aqua platforms (500 m spatial resolution and daily revisit cycle). We implem…

010504 meteorology & atmospheric sciencesComputer science0208 environmental biotechnologyMultispectral imageSoil Science02 engineering and technology01 natural sciencesArticleComputers in Earth SciencesImage resolution0105 earth and related environmental sciencesRemote sensingPropagation of uncertaintyNoise (signal processing)GeologyKalman filterData fusionSensor fusion020801 environmental engineeringMODIS13. Climate actionScalabilityGap fillingKalman filterLandsatSmoothingSmoothingRemote Sensing of Environment
researchProduct

A multisensor fusion approach to improve LAI time series

2011

International audience; High-quality and gap-free satellite time series are required for reliable terrestrial monitoring. Moderate resolution sensors provide continuous observations at global scale for monitoring spatial and temporal variations of land surface characteristics. However, the full potential of remote sensing systems is often hampered by poor quality or missing data caused by clouds, aerosols, snow cover, algorithms and instrumentation problems. A multisensor fusion approach is here proposed to improve the spatio-temporal continuity, consistency and accuracy of current satellite products. It is based on the use of neural networks, gap filling and temporal smoothing techniques. …

010504 meteorology & atmospheric sciencesMeteorologytélédétectionsatellite0211 other engineering and technologiesSoil Scienceréseau neuronal02 engineering and technology01 natural sciencessuivi de culturesInstrumentation (computer programming)Computers in Earth SciencesLeaf area index021101 geological & geomatics engineering0105 earth and related environmental sciencesRemote sensingVegetationGeologyVegetationData fusionLAI time seriesSensor fusionMissing dataLAI time series;Vegetation;Modis;Temporal smoothing;Gap filling;Data fusionqualité des données13. Climate actionAutre (Sciences de l'ingénieur)Gap filling[SDE]Environmental SciencesEnvironmental scienceSatelliteModisTemporal smoothingScale (map)Smoothing
researchProduct

Green LAI Mapping and Cloud Gap-Filling Using Gaussian Process Regression in Google Earth Engine

2021

For the last decade, Gaussian process regression (GPR) proved to be a competitive machine learning regression algorithm for Earth observation applications, with attractive unique properties such as band relevance ranking and uncertainty estimates. More recently, GPR also proved to be a proficient time series processor to fill up gaps in optical imagery, typically due to cloud cover. This makes GPR perfectly suited for large-scale spatiotemporal processing of satellite imageries into cloud-free products of biophysical variables. With the advent of the Google Earth Engine (GEE) cloud platform, new opportunities emerged to process local-to-planetary scale satellite data using advanced machine …

Earth observationGoogle Earth Engine (GEE); Gaussian process regression (GPR); machine learning; Sentinel-2; gap filling; leaf area index (LAI)010504 meteorology & atmospheric sciencesComputer scienceScienceleaf area index (LAI)0211 other engineering and technologiesCloud computing02 engineering and technologycomputer.software_genre01 natural sciencesKrigingGaussian process regression (GPR)021101 geological & geomatics engineering0105 earth and related environmental sciencesPixelbusiness.industryQGoogle Earth Engine (GEE)machine learningKernel (image processing)Ground-penetrating radarGeneral Earth and Planetary SciencesData miningSentinel-2Scale (map)businesscomputergap fillingLevel of detailRemote Sensing; Volume 13; Issue 3; Pages: 403
researchProduct

Filling in long gap sequences by performing jointly EOF and FDA

2012

In this paper the EOF methodology is performed jointly with the FDA approach on a spatiotemporal multivariate data set with the aim to fill in missing values as accurately as possible when long gap sequences occur. Simulated data sets, containing ”artificial” gaps, are considered in order to test the performance of two proposed procedures; in the first one, observed data are reconstructed by EOF and then converted into functional ones; in the second one, observed data are transformed into functional ones and then EOF reconstruction is applied. By comparing some performance indicators computed for the two procedures, it is shown that a pre-processing of data by FDA, followed by the EOF, may …

FDA EOF gap filling
researchProduct

Integrating Domain Knowledge in Data-Driven Earth Observation With Process Convolutions

2022

The modelling of Earth observation data is a challenging problem, typically approached by either purely mechanistic or purely data-driven methods. Mechanistic models encode the domain knowledge and physical rules governing the system. Such models, however, need the correct specification of all interactions between variables in the problem and the appropriate parameterization is a challenge in itself. On the other hand, machine learning approaches are flexible data-driven tools, able to approximate arbitrarily complex functions, but lack interpretability and struggle when data is scarce or in extrapolation regimes. In this paper, we argue that hybrid learning schemes that combine both approa…

FOS: Computer and information sciencesComputer Science - Machine LearningEarth observationAdvanced microwave scanning radiometer-2 (AMSR-2)moderate resolution imaging spectroradiometer (MODIS)Computer scienceleaf area index (LAI)0211 other engineering and technologiesExtrapolationMachine Learning (stat.ML)02 engineering and technologycomputer.software_genreMachine Learning (cs.LG)Data-drivenConvolutionsymbols.namesakeadvanced scatterometer (ASCAT)Statistics - Machine Learningordinary differential equation (ODE)Electrical and Electronic EngineeringGaussian processsoil moisture and ocean salinity (SMOS)021101 geological & geomatics engineeringInterpretabilityForcing (recursion theory)machine learning (ML)soil moisture (SM)time series analysisgaussian process (GP)symbolsGeneral Earth and Planetary SciencesDomain knowledgeData mininggap fillingphysicscomputerfraction of absorbed photosynthetically active radiation (faPAR)IEEE Transactions on Geoscience and Remote Sensing
researchProduct

EOFs for gap filling in multivariate air quality data: a FDA approach

2010

Missing values are a common concern in spatiotemporal data sets. During recent years a great number of methods have been developed for gap filling. One of the emerging approaches is based on the Empirical Orthogonal Function (EOF) methodology, applied mainly on raw and univariate data sets presenting irregular missing patterns. In this paper EOF is carried out on a multivariate space-time data set, related to concentrations of pollutants recorded at different sites, after denoising raw data by FDA approach. Some performance indicators are computed on simulated incomplete data sets with also long gaps in order to show that the EOF reconstruction appears to be an improved procedure especially…

Settore SECS-S/01 - StatisticaFDA EOF missing data gap filling
researchProduct