Accuracy of selected methods of measurement of tree heights

Abstract

In this paper, the possibilities of using stereophotogrammetry methods for measurements using unmanned aerial vehicles (UAVs) for the conditions of a mature pine stand with uneven density are examined. Here, we carried out a comparison of measurements using altimeters and remote sensing data collected with a UAV optical camera. In particular, the height of growing trees was estimated by three different field-based devices and applying the three methods of data collection and processing with UAVs. Specifically, one method implied the direct measurements using on-board UAV equipment. The following two methods are based on the data provided by the stereophotogrammetrical approach, while the aerial images for that were collected using a UAV optical camera. In particular, there was a modeling of the points cloud from one-sided vertical shooting of sample trees and determination of height of trees from digital canopy height model (CHM) from data of aerial photography of horizontal spans over a stand. Our investigation confirmed the highest accuracy of laser measuring tools among the ground measuring devices used in research. Respective value of the average random measurement error was less than 3 % (0.88 m). Among the results obtained from the analysis of the original data collected by UAVs, the best method was to utilize the CHM, namely, the average random error was less than 2% (0.64 m). This exceeds the accuracy of laser altimeter measurements 33 %. Thus, this method of measuring height in pine stands meets the standards of accuracy in determining the height for production assessment, according to the “Inventory guidelines for the forest fund of Ukraine”, and can be used for survey, inventory, forest management and other works related to forestry and monitoring the changes in forest ecosystems.