Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery
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...
| Main Authors: | , , , , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/20/17/4802 |
| _version_ | 1797555549885169664 |
|---|---|
| author | Wellington Castro José Marcato Junior Caio Polidoro Lucas Prado Osco Wesley Gonçalves Lucas Rodrigues Mateus Santos Liana Jank Sanzio Barrios Cacilda Valle Rosangela Simeão Camilo Carromeu Eloise Silveira Lúcio André de Castro Jorge Edson Matsubara |
| author_facet | Wellington Castro José Marcato Junior Caio Polidoro Lucas Prado Osco Wesley Gonçalves Lucas Rodrigues Mateus Santos Liana Jank Sanzio Barrios Cacilda Valle Rosangela Simeão Camilo Carromeu Eloise Silveira Lúcio André de Castro Jorge Edson Matsubara |
| author_sort | Wellington Castro |
| collection | DOAJ |
| description | 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 <i>Panicum maximum</i> 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. |
| first_indexed | 2024-03-10T16:49:07Z |
| format | Article |
| id | doaj.art-c69227f110e941249022525a9b5d3464 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-10T16:49:07Z |
| publishDate | 2020-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-c69227f110e941249022525a9b5d34642023-11-20T11:21:58ZengMDPI AGSensors1424-82202020-08-012017480210.3390/s20174802Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB ImageryWellington Castro0José Marcato Junior1Caio Polidoro2Lucas Prado Osco3Wesley Gonçalves4Lucas Rodrigues5Mateus Santos6Liana Jank7Sanzio Barrios8Cacilda Valle9Rosangela Simeão10Camilo Carromeu11Eloise Silveira12Lúcio André de Castro Jorge13Edson Matsubara14Faculty of Computer Science, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilFaculty of Computer Science, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilFaculty of Engineering, Architecture and Urbanism, University of Western São Paulo, Presidente Prudente 19067175, SP, BrazilFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilFaculty of Computer Science, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilEmbrapa Beef Cattle, Brazilian Agricultural Research Corporation, Campo Grande 79106550, MS, BrazilEmbrapa Beef Cattle, Brazilian Agricultural Research Corporation, Campo Grande 79106550, MS, BrazilEmbrapa Beef Cattle, Brazilian Agricultural Research Corporation, Campo Grande 79106550, MS, BrazilEmbrapa Beef Cattle, Brazilian Agricultural Research Corporation, Campo Grande 79106550, MS, BrazilEmbrapa Beef Cattle, Brazilian Agricultural Research Corporation, Campo Grande 79106550, MS, BrazilEmbrapa Beef Cattle, Brazilian Agricultural Research Corporation, Campo Grande 79106550, MS, BrazilFaculty of Engineering, Architecture and Urbanism and Geography, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilEmbrapa Instrumentation, São Carlos 13560970, SP, BrazilFaculty of Computer Science, Federal University of Mato Grosso do Sul, Campo Grande 79070900, MS, BrazilMonitoring 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 <i>Panicum maximum</i> 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.https://www.mdpi.com/1424-8220/20/17/4802Convolutional Neural Networkbiomass yielddata augmentationphenotyping |
| spellingShingle | Wellington Castro José Marcato Junior Caio Polidoro Lucas Prado Osco Wesley Gonçalves Lucas Rodrigues Mateus Santos Liana Jank Sanzio Barrios Cacilda Valle Rosangela Simeão Camilo Carromeu Eloise Silveira Lúcio André de Castro Jorge Edson Matsubara Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery Sensors Convolutional Neural Network biomass yield data augmentation phenotyping |
| title | Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery |
| title_full | Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery |
| title_fullStr | Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery |
| title_full_unstemmed | Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery |
| title_short | Deep Learning Applied to Phenotyping of Biomass in Forages with UAV-Based RGB Imagery |
| title_sort | deep learning applied to phenotyping of biomass in forages with uav based rgb imagery |
| topic | Convolutional Neural Network biomass yield data augmentation phenotyping |
| url | https://www.mdpi.com/1424-8220/20/17/4802 |
| work_keys_str_mv | AT wellingtoncastro deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT josemarcatojunior deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT caiopolidoro deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT lucaspradoosco deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT wesleygoncalves deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT lucasrodrigues deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT mateussantos deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT lianajank deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT sanziobarrios deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT cacildavalle deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT rosangelasimeao deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT camilocarromeu deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT eloisesilveira deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT lucioandredecastrojorge deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery AT edsonmatsubara deeplearningappliedtophenotypingofbiomassinforageswithuavbasedrgbimagery |