The Neural Network Algorithm for Estimating Plantation Forest Leaf Area Index based on ENVISAT/ASAR
Based on Neural Network (NN) algorithm,this paper estimated the Leaf Area Index of plantation forests in HEIHE area northwest china using the ENVISAT/ASAR data. Firstly,the relationship between ENVISAT/ASAR microwave backscattering coefficients (σ°) with different polarization, incidence angle and Leaf Area Index (LAI) of White Poplar and Desert Date planted forests were analyzed. The study showed that the homogeneity of plantation forests was the primary factor influencing the relationship. And there were significant difference in the relationship in different incidence angle conditions. Based on the a-nalysis above,the NN algorithm for LAI retrieval was designed using ENVISAT/ASAR with different polarization, incidence angle as input parameters. The study compared the ground measured and predicted LAI for validated data,training data and all data. The determination coefficient reached 0. 61,0. 91 and 0. 82 respectively. And the results showed that there was great feasibility to estimate LAI of plantation forests by NN algorithm using ENVISAT/ASAR data.
Leaf Area Index (LAI)SARBackscattering coefficientENVISAT ASARNeural network
NETL
NSTL
万方数据
维普
