Height estimation of a clonal Eucalyptus test at different ages using low-altitude aerial imaging

Nenhuma Miniatura disponível

Data

2022-01-01

Autores

Martello, Mauricio
Alvares, Clayton Alcarde [UNESP]
Tavares, Tiago Rodrigues
Fiorio, Peterson Ricardo
Campoe, Otavio Camargo [UNESP]
Carneiro, Rafaela Lorenzato

Título da Revista

ISSN da Revista

Título de Volume

Editor

Ipef-inst Pesquisas Estudos Florestais

Resumo

This study evaluated the performance of the estimation of Eucalyptus height using aerial images obtained with a remotely piloted aircraft (RPA). The study area is part of the Cooperative Program on Tolerance of Clonal Eucalyptus by Thermal and Biotic Water Stresses (TECHS-IPEF). Four aerial image acquisitions were carried out at different stages of plant development, evaluating 15 experimental plots, each with a different clone of Eucalyptus. The plots contained 5 rows, spaced at 2.3 m, with 4 plants in each row (also spaced at 2.3 m), totaling 20 trees and 5.06 m(2) per plant. The image acquisition was done using aircrafts with digital cameras that had a sensitivity in the visible spectral region (RGB). The images were synchronized with a global navigation satellite system (GNSS) receiver for geo-referencing the information. The acquisition of stereoscopic images allowed the generation of three-dimensional models that were used to estimate the height of each eucalyptus plant, in each of the four image collections. The heights estimated with the RPA images were compared with height measurements performed in the field by means of linear regressions. The results showed an estimation performance with a coefficient of determination (R-2) value higher than 0.75 and root mean square error (RMSE) lower than 3.94 m. It is concluded that height estimation using aerial images has a high potential for monitoring and evaluation Eucalyptus clones, and can be a useful tool to support the management of these areas.

Descrição

Palavras-chave

Remote sensing, Geo-processing, Remotely piloted aircraft

Como citar

Scientia Forestalis. Piracicaba: Ipef-inst Pesquisas Estudos Florestais, v. 50, 13 p., 2022.

Coleções