Influence of noise on decay predictions in standing trees

Abstract

In the present study, the stochastic process, termed the Vasicek, have been implemented to analyze propagation velocity data of stress wave that diffuses through the wood of decayed and sound stems in standing trees. Stochastic processes offer the possibility of capturing important distributional differences between stress wave velocity datasets of decayed and sound trees and make the detecting rule flexible to internal decay in standing trees. It is shown that the Vasicek type stochastic process is a possible candidate for statistical classification of waveforms. Rigorous quantile regression analysis is based on the conditional probability density functions whose are fitted for stress wave velocity datasets of sound and decayed trees. New developed detector rules derived from stochastic differential equations may be used as an initial tool for detecting internal decay in standing trees.