An Emulator Toolbox to Approximate Radiative Transfer Models with Statistical Learning

Remote Sensing

Juan Pablo Rivera²  Jochem Verrelst²  Jose Gómez-Dans³  Jordi Muñoz-Marí²  José Moreno²  Gustau Camps-Valls²  Clement Atzberger¹ 

[1]Image Processing Laboratory (IPL), Parc Científic, Universitat de València, 46980 Paterna, València, Spain

[2]

[3]Image Processing Laboratory (IPL), Parc Científic, Universitat de València, 46980 Paterna, València, Spain

[4] E-Mails:

[5]Department of Geography, University College London, London WC1E 6BT, UK

[6] E-Mail:

关键词: emulator;    machine learning;    radiative transfer models;    multi-output;    ARTMO;    GUI toolbox;    FLEX;    fluorescence;   

DOI  :  10.3390/rs70709347

来源: mdpi

【 摘 要 】

Physically-based radiative transfer models (RTMs) help in understanding the processes occurring on the Earth’s surface and their interactions with vegetation and atmosphere. When it comes to studying vegetation properties, RTMs allows us to study light interception by plant canopies and are used in the retrieval of biophysical variables through model inversion. However, advanced RTMs can take a long computational time, which makes them unfeasible in many real applications. To overcome this problem, it has been proposed to substitute RTMs through so-called emulators. Emulators are statistical models that approximate the functioning of RTMs. Emulators are advantageous in real practice because of the computational efficiency and excellent accuracy and flexibility for extrapolation. We hereby present an “Emulator toolbox” that enables analysing multi-output machine learning regression algorithms (MO-MLRAs) on their ability to approximate an RTM. The toolbox is included in the free-access ARTMO’s MATLAB suite for parameter retrieval and model inversion and currently contains both linear and non-linear MO-MLRAs, namely partial least squares regression (PLSR), kernel ridge regression (KRR) and neural networks (NN). These MO-MLRAs have been evaluated on their precision and speed to approximate the soil vegetation atmosphere transfer model SCOPE (Soil Canopy Observation, Photochemistry and Energy balance). SCOPE generates, amongst others, sun-induced chlorophyll fluorescence as the output signal. KRR and NN were evaluated as capable of reconstructing fluorescence spectra with great precision. Relative errors fell below 0.5% when trained with 500 or more samples using cross-validation and principal component analysis to alleviate the underdetermination problem. Moreover, NN reconstructed fluorescence spectra about 50-times faster and KRR about 800-times faster than SCOPE. The Emulator toolbox is foreseen to open new opportunities in the use of advanced RTMs, in which both consistent physical assumptions and data-driven machine learning algorithms live together.

【 授权许可】

CC BY   
© 2015 by the authors; licensee MDPI, Basel, Switzerland