A neural network-based scheme coupled with the RPV model inversion package

Abstract A hybrid scheme for the inversion of the Rahman–Pinty–Verstraete (RPV) model is presented. It combines the inversion technique described by Lavergne et al. (Lavergne, T., Kaminski, T., Pinty, B., Taberner, M., Gobron, N., Verstraete, M.M., Vossbeck, M., Widlowski, J.L., Giering, R. (2007). Application to MISR land products of an RPV model inversion package using adjoint and Hessian codes. Remote Sensing of Environment, 107, 362–375.) and a multilayer backpropagation feedforward neural network. The RPV inversion package is applied to a sample set of pixels within the satellite scene. Subsequently the pairs of bidirectional reflectance factors (BRF) and model parameters estimated from the sample set of pixels are used to train the neural network. Since the mathematical formulation of the RPV model is embedded in these training data variables, the neural network can efficiently retrieve the model parameters for the whole satellite scene. This scheme has been tested for a MISR L2 BRF scene, MISR L1B2-derived BRF data corresponding to two different dates and a mosaic of MISR L2 BRF scenes acquired over Southern Africa covering a large extent of Miombo woodland. The results show this strategy retrieves the RPV model parameters and uncertainties with high accuracy and considerable speed over large areas.

[1]  Nadine Gobron,et al.  Uniqueness of multiangular measurements. I. An indicator of subpixel surface heterogeneity from MISR , 2002, IEEE Trans. Geosci. Remote. Sens..

[2]  Andreas Griewank,et al.  Evaluating derivatives - principles and techniques of algorithmic differentiation, Second Edition , 2000, Frontiers in applied mathematics.

[3]  W. K. Hastings,et al.  Monte Carlo Sampling Methods Using Markov Chains and Their Applications , 1970 .

[4]  D. Marquardt.  An Algorithm for Least-Squares Estimation of Nonlinear Parameters , 1963 .

[5]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[6]  Gareth Roberts,et al.  Vegetation structure characteristics and relationships of Kalahari woodlands and savannas , 2004 .

[7]  Ranga B. Myneni,et al.  Time‐series validation of MODIS land biophysical products in a Kalahari woodland, Africa , 2005 .

[8]  F. Baret,et al.  Neural network estimation of LAI, fAPAR, fCover and LAI×Cab, from top of canopy MERIS reflectance data : Principles and validation , 2006 .

[9]  H. Rahman,et al.  Coupled surface-atmosphere reflectance (CSAR) model: 2. Semiempirical surface model usable with NOAA advanced very high resolution radiometer data , 1993 .

[10]  W. Lucht,et al.  Considerations in the parametric modeling of BRDF and albedo from multiangular satellite sensor observations , 2000 .

[11]  E. Satterfield,et al.  Cover: Southern Africa, 2 February 2002 to 16 May 2002, Moderate Resolution Imaging Spectroradiometer (MODIS) 500 m land surface reflectance , 2005 .

[12]  S. Ustin,et al.  Multi-temporal vegetation canopy water content retrieval and interpretation using artificial neural networks for the continental USA , 2008 .

[13]  O. Hagolle,et al.  LAI, fAPAR and fCover CYCLOPES global products derived from VEGETATION: Part 1: Principles of the algorithm , 2007 .

[14]  Thomas Kaminski,et al.  Recipes for adjoint code construction , 1998, TOMS.

[15]  Peng Gong,et al.  Land cover assessment with MODIS imagery in southern African Miombo ecosystems , 2005 .

[16]  Peng Gong,et al.  Inverting a canopy reflectance model using a neural network , 1999 .

[17]  Nadine Gobron,et al.  A new inversion scheme for the RPV model , 2002 .

[18]  Michael T. Manry,et al.  Attributes of neural networks for extracting continuous vegetation variables from optical and radar , 1998 .

[19]  A. Strahler,et al.  Forward and inverse modelling of canopy directional reflectance using a neural network , 1998 .

[20]  A. Strahler,et al.  A review of reflectance nomenclature used in remote sensing , 2000 .

[21]  F. Baret,et al.  Validation of neural net techniques to estimate canopy biophysical variables from remote sensing data , 2002 .

[22]  Nadine Gobron,et al.  Characterization of surface heterogeneity detected at the MISR/TERRA subpixel scale , 2001 .

[23]  R. Giering,et al.  Application to MISR land products of an RPV model inversion package using adjoint and Hessian codes , 2007 .

[24]  Gregory P. Asner,et al.  Ecological Research Needs from Multiangle Remote Sensing Data , 1998 .

[25]  S. Liang.  Quantitative Remote Sensing of Land Surfaces , 2003 .

[26]  D. Diner,et al.  MISR: A multiangle imaging spectroradiometer for geophysical and climatological research from Eos , 1989 .

[27]  A. Tarantola.  Inverse problem theory : methods for data fitting and model parameter estimation , 1987 .

[28]  Peng Gong,et al.  Inverting a canopy reflectance model using an artificial neural network , 1995, Remote Sensing.

[29]  Daniel Gómez,et al.  Tree density estimation in a tropical woodland ecosystem with multiangular MISR and MODIS data , 2008 .