Retrieve leaf area index from HJ-CCD image based on support vector regression and physical model

Jingjing Pan; Hua Yang; Wei He Sr.; Peipei Xu

doi:10.1117/12.2029061

16 October 2013 Retrieve leaf area index from HJ-CCD image based on support vector regression and physical model

Jingjing Pan, Hua Yang, Wei He Sr., Peipei Xu

Proceedings Volume 8887, Remote Sensing for Agriculture, Ecosystems, and Hydrology XV; 88871R (2013) https://doi.org/10.1117/12.2029061
Event: SPIE Remote Sensing, 2013, Dresden, Germany

Abstract

Many vegetation biophysical parameters, e.g. Leaf Area Index, have nonlinear relationships with spectral reflectance, while Support Vector Regression (SVR) has fantastic nonlinear fitting ability owing to kernel inner product and the sparseness of Support Vectors (SVs). Currently, most of the LAI inversion methods over Huanjing optical satellites (HJ) are based on empirical relationship between LAI and spectral index, which have some limitations. In this paper, we developed an algorithm combining SVR and physical model-PROSAIL to retrieve LAI over HJ-CCD image. The model adopted a new HJ vegetation index (HJVI), which can lessen saturation on high LAI domain. Experiments over simulations generated by PROSAIL model proved the new algorithm’s good performance on noising resisting and effectiveness of HJVI. Finally, we applied the algorithm on a HJ1B CCD2 image and validated it with the field measured data in Xinxiang, Henan Province, China. The RMSE of 0.5230 indicated the applicability of the SVR-based physical method over HJ data.

Citation Download Citation

Jingjing Pan, Hua Yang, Wei He Sr., and Peipei Xu "Retrieve leaf area index from HJ-CCD image based on support vector regression and physical model", Proc. SPIE 8887, Remote Sensing for Agriculture, Ecosystems, and Hydrology XV, 88871R (16 October 2013); https://doi.org/10.1117/12.2029061

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