|
|
| Acesso ao texto completo restrito à biblioteca da Embrapa Agricultura Digital. Para informações adicionais entre em contato com cnptia.biblioteca@embrapa.br. |
Registro Completo |
Biblioteca(s): |
Embrapa Agricultura Digital. |
Data corrente: |
07/02/2020 |
Data da última atualização: |
17/04/2020 |
Tipo da produção científica: |
Artigo em Periódico Indexado |
Autoria: |
SANTOS, T. T.; SOUZA, L. L. de; SANTOS, A. A. dos; AVILA, S. |
Afiliação: |
THIAGO TEIXEIRA SANTOS, CNPTIA; LEONARDO L. DE SOUZA, IC/Unicamp; ANDREZA A. DOS SANTOS, IC/Unicamp; SANDRA AVILA, IC/Unicamp. |
Título: |
Grape detection, segmentation, and tracking using deep neural networks and three-dimensional association. |
Ano de publicação: |
2020 |
Fonte/Imprenta: |
Computers and Electronics in Agriculture, v. 170, p. 1-17, Mar. 2020. |
DOI: |
https://doi.org/10.1016/j.compag.2020.105247 |
Idioma: |
Inglês |
Notas: |
Article 105247. |
Conteúdo: |
Agricultural applications such as yield prediction, precision agriculture and automated harvesting need systems able to infer the crop state from low-cost sensing devices. Proximal sensing using affordable cameras combined with computer vision has seen a promising alternative, strengthened after the advent of convolutional neural networks (CNNs) as an alternative for challenging pattern recognition problems in natural images. Considering fruit growing monitoring and automation, a fundamental problem is the detection, segmentation and counting of individual fruits in orchards. Here we show that for wine grapes, a crop presenting large variability in shape, color, size and compactness, grape clusters can be successfully detected, segmented and tracked using state-of-the-art CNNs. In a test set containing 408 grape clusters from images taken on a trellis-system based vineyard, we have reached an F1-score up to 0.91 for instance segmentation, a fine separation of each cluster from other structures in the image that allows a more accurate assessment of fruit size and shape. We have also shown as clusters can be identified and tracked along video sequences recording orchard rows. We also present a public dataset containing grape clusters properly annotated in 300 images and a novel annotation methodology for segmentation of complex objects in natural images. The presented pipeline for annotation, training, evaluation and tracking of agricultural patterns in images can be replicated for different crops and production systems. It can be employed in the development of sensing components for several agricultural and environmental applications. MenosAgricultural applications such as yield prediction, precision agriculture and automated harvesting need systems able to infer the crop state from low-cost sensing devices. Proximal sensing using affordable cameras combined with computer vision has seen a promising alternative, strengthened after the advent of convolutional neural networks (CNNs) as an alternative for challenging pattern recognition problems in natural images. Considering fruit growing monitoring and automation, a fundamental problem is the detection, segmentation and counting of individual fruits in orchards. Here we show that for wine grapes, a crop presenting large variability in shape, color, size and compactness, grape clusters can be successfully detected, segmented and tracked using state-of-the-art CNNs. In a test set containing 408 grape clusters from images taken on a trellis-system based vineyard, we have reached an F1-score up to 0.91 for instance segmentation, a fine separation of each cluster from other structures in the image that allows a more accurate assessment of fruit size and shape. We have also shown as clusters can be identified and tracked along video sequences recording orchard rows. We also present a public dataset containing grape clusters properly annotated in 300 images and a novel annotation methodology for segmentation of complex objects in natural images. The presented pipeline for annotation, training, evaluation and tracking of agricultural patterns in images can be replicate... Mostrar Tudo |
Palavras-Chave: |
Aprendizado profundo; Convolutional neural networks; Deep learning; Detecção de fruta; Detecção de uva; Fruit detection; Previsão de rendimento; Redes neurais convolucionais; Redes neurais profundas; Visão computacional; Yield prediction. |
Thesagro: |
Agricultura; Fruta; Uva. |
Thesaurus Nal: |
Agriculture; Computer vision; Neural networks. |
Categoria do assunto: |
X Pesquisa, Tecnologia e Engenharia |
Marc: |
LEADER 02844naa a2200385 a 4500 001 2119989 005 2020-04-17 008 2020 bl uuuu u00u1 u #d 024 7 $ahttps://doi.org/10.1016/j.compag.2020.105247$2DOI 100 1 $aSANTOS, T. T. 245 $aGrape detection, segmentation, and tracking using deep neural networks and three-dimensional association.$h[electronic resource] 260 $c2020 500 $aArticle 105247. 520 $aAgricultural applications such as yield prediction, precision agriculture and automated harvesting need systems able to infer the crop state from low-cost sensing devices. Proximal sensing using affordable cameras combined with computer vision has seen a promising alternative, strengthened after the advent of convolutional neural networks (CNNs) as an alternative for challenging pattern recognition problems in natural images. Considering fruit growing monitoring and automation, a fundamental problem is the detection, segmentation and counting of individual fruits in orchards. Here we show that for wine grapes, a crop presenting large variability in shape, color, size and compactness, grape clusters can be successfully detected, segmented and tracked using state-of-the-art CNNs. In a test set containing 408 grape clusters from images taken on a trellis-system based vineyard, we have reached an F1-score up to 0.91 for instance segmentation, a fine separation of each cluster from other structures in the image that allows a more accurate assessment of fruit size and shape. We have also shown as clusters can be identified and tracked along video sequences recording orchard rows. We also present a public dataset containing grape clusters properly annotated in 300 images and a novel annotation methodology for segmentation of complex objects in natural images. The presented pipeline for annotation, training, evaluation and tracking of agricultural patterns in images can be replicated for different crops and production systems. It can be employed in the development of sensing components for several agricultural and environmental applications. 650 $aAgriculture 650 $aComputer vision 650 $aNeural networks 650 $aAgricultura 650 $aFruta 650 $aUva 653 $aAprendizado profundo 653 $aConvolutional neural networks 653 $aDeep learning 653 $aDetecção de fruta 653 $aDetecção de uva 653 $aFruit detection 653 $aPrevisão de rendimento 653 $aRedes neurais convolucionais 653 $aRedes neurais profundas 653 $aVisão computacional 653 $aYield prediction 700 1 $aSOUZA, L. L. de 700 1 $aSANTOS, A. A. dos 700 1 $aAVILA, S. 773 $tComputers and Electronics in Agriculture$gv. 170, p. 1-17, Mar. 2020.
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Agricultura Digital (CNPTIA) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
Registros recuperados : 1 | |
1. | | BAUMGRATZ, E. I.; MERA, C. M. P. de; FIORIN, J. E.; CASTRO, N. L. M. de; CASTRO, R. de. Produção de trigo: a decisão por análise econômico-financeira. Revista de Política Agrícola, Brasília, DF, ano 26, n. 3, p. 8-21, jul./ago./set. 2017. Título em inglês: Wheat production: the decision for economic-financial analysis.Biblioteca(s): Embrapa Unidades Centrais. |
| |
Registros recuperados : 1 | |
|
Expressão de busca inválida. Verifique!!! |
|
|