Estimation of aboveground biomass of potato based on ground hyperspectral

Abstract

Biomass is an important indicator of crop population characteristics and growth monitoring. Rapid and accurate monitoring of crop biomass is important for precise management of farmland. The spectral indices of the combination of any two bands of 350~2500nm were obtained that have good correlation with biomass were screened out through correlation analysis. At the same time, they were as input variables of biomass estimation models. Above-biomass of potato estimation models were established with partial least squares regression (PLSR), multiple linear regression (MLR) and random forest (RF). The result showed the potato tuber formation period and the tuber growth period, the combination index using the PLSR method to construct the potato biomass estimation model is higher, the starch accumulation period and the mature period, the combination index using MLR method to construct the biomass estimation model is high, can be better to realize the potato biomass estimation