Distributed reflectance model mining of leaf nitrogen content by using gene expression programming

摘要

Estimating dynamically for leaf nitrogen concentration is an important subject in the studies on crop monitoring. The traditional regression methods including linear regression, partial least squares regression, support vector machine regression and random forest regression depended on a priori knowledge and many subjective factors. Moreover, these methods have high time complexity and low computational efficiency for complex and high-dimensional hyperspectral data. In order to better find reflectance model of LNC for complex and high-dimensional hyperspectral data, this paper presents distributed reflectance model mining of leaf nitrogen content by using gene expression programming (DRMMLNC-GEP) which combined with GEP and grid service. The comparative results show that the DRMMLNC-GEP outperforms all other algorithms on the average time-consumption, value of R-square and prediction accuracy. Meanwhile, experimental results also show that with the increasing of datasets size, DRMMLNC-GEP demonstrates good speed-up ratio and scale-up ratio too.