Portal de Periódicos da CAPES

Carta Acesso aberto Produção Nacional Revisado por pares

BIBTEX RIS

Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery

2020; Multidisciplinary Digital Publishing Institute; Volume: 20; Issue: 17 Linguagem: Inglês

10.3390/s20174802

1424-8220

Wellington V. M. de Castro, José Marcato, Caio Polidoro, Lucas Prado Osco, Wesley Nunes Gonçalves, Lucas de Souza Rodrigues, M. F. Santos, Liana Jank, Sanzio Carvalho Lima Barrios, Cacilda Borges do Valle, R. M. S. Resende, Camilo Carromeu, Eloise Silveira, Lúcio André de Castro Jorge, Edson Takashi Matsubara,

Spectroscopy and Chemometric Analyses

Monitoring biomass of forages in experimental plots and livestock farms is a time-consuming, expensive, and biased task. Thus, non-destructive, accurate, precise, and quick phenotyping strategies for biomass yield are needed. To promote high-throughput phenotyping in forages, we propose and evaluate the use of deep learning-based methods and UAV (Unmanned Aerial Vehicle)-based RGB images to estimate the value of biomass yield by different genotypes of the forage grass species Panicum maximum Jacq. Experiments were conducted in the Brazilian Cerrado with 110 genotypes with three replications, totaling 330 plots. Two regression models based on Convolutional Neural Networks (CNNs) named AlexNet and ResNet18 were evaluated, and compared to VGGNet—adopted in previous work in the same thematic for other grass species. The predictions returned by the models reached a correlation of 0.88 and a mean absolute error of 12.98% using AlexNet considering pre-training and data augmentation. This proposal may contribute to forage biomass estimation in breeding populations and livestock areas, as well as to reduce the labor in the field.