Rice biomass retrieval from multitemporal ground-based scatterometer data and RADARSAT-2 images using neural networks

Mingquan Jia; Ling Tong; Yan Chen; Yong Wang; Yuanzhi Zhang

doi:10.1117/1.JRS.7.073509

6 September 2013 Rice biomass retrieval from multitemporal ground-based scatterometer data and RADARSAT-2 images using neural networks

Mingquan Jia, Ling Tong, Yan Chen, Yong Wang, Yuanzhi Zhang

Author Affiliations +

Journal of Applied Remote Sensing, Vol. 7, Issue 1, 073509 (September 2013). https://doi.org/10.1117/1.JRS.7.073509

Abstract

A neural network (NN) algorithm to invert biomass of rice plants using quad-polarization radar datasets of ground-based scatterometer and spaceborne RADARSAT-2 has been studied. The NN is trained with pairs of multipolarization radar backscattering and biomass data. The backscattering data are simulated from a Monte Carlo backscatter model that uses the outputs from a growth model of the rice plant. The growth model is developed from the plant data collected in growing cycles of several years. In addition to producing parameters needed by the backscatter model, the growth model outputs the biomass value of the plant. Multipolarization data collected by a ground-based scatterometer at eight stages during the 2012 growing cycle are input to the NN to invert biomass. Satisfactory results are obtained due to a small root mean squared error (RMSE) of 0.477 kg/m ² and a high correlation coefficient of 0.989 when the inverted and measured biomass values are compared. Finally, RADARSAT-2 synthetic aperture radar images acquired on four different dates during the 2012 growth period are analyzed to delineate rice paddies within the study area and to invert biomass using the NN. Inversion results from the delineated rice paddies are encouraging because the RMSE is 0.582 kg/m ² and correlation coefficient is 0.983.

Citation Download Citation

Mingquan Jia, Ling Tong, Yan Chen, Yong Wang, and Yuanzhi Zhang "Rice biomass retrieval from multitemporal ground-based scatterometer data and RADARSAT-2 images using neural networks," Journal of Applied Remote Sensing 7(1), 073509 (6 September 2013). https://doi.org/10.1117/1.JRS.7.073509

Published: 6 September 2013

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

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.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available