Mapeamento da relação altura-diâmentro utilizando vector machine

Abstract

Height determination is an onerous and error-prone operation, more recently with the evolution of remote sensing technologies, especially LiDAR, obtaining the height variable is done directly and immediately. In this work, we sought to evaluate the diameter estimates at 1.30 m height from the Support Vector Machine (SMV) analysis tool. Two criteria were used to divide the database for training and validation, with Training 1 being done using data from the years 2012, 2013 and 2014 for training and the year 2015 for validation, and Training 2 considering a random division of data. Statistica software for SMV training was used. After the processing, the Bias, RQME, correlation and mean percent error statistics were calculated; dispersion charts were generated, considering the estimated and observed values; and histogram of residues. It was possible to conclude that the SMV can be used to estimate diameters at a height of 1.30 m. and that for more accurate estimates it is necessary that the training data be representative of the study population.