Sex type determination in papaya seeds and leaves using near infrared spectroscopy combined with multivariate techniques and machine learning
Nenhuma Miniatura disponível
Data
2022-02-01
Autores
Fernandes, Thiago Feliph Silva [UNESP]
Silva, Raíssa Vanessa de Oliveira
Freitas, Daniel Lucas Dantas de
Sanches, Alex Guimarães [UNESP]
Silva, Maryelle Barros da [UNESP]
Cunha Júnior, Luis Carlos
Lima, Kássio Gomes de
Teixeira, Gustavo Henrique de Almeida [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Resumo
Papaya trees (Carica papaya L.) can bear female, hermaphrodite, and male flowers. However, only the hermaphrodite type produces elongate fruit required and appreciated by the consumer market. Sex type determination is carried out based on the plant phenotype after planting the seedlings, increasing the production costs. In this regard, the objective of this study was to verify the feasibility of using near infrared spectroscopy (NIR) as a non-destructive method for sexing papaya trees. The NIR spectra were collected using seeds and leaves of the respective seedlings of the cultivars ‘T2′, ‘Formosa’, and ‘Calimosa’ (Formosa group), and ‘THB’ and ‘Ouro’ (Solo group). By using the seeds, it was possible to obtain a F-score value of 0.81 for the external validation set applying the principal component analysis and quadratic discriminant analysis (PCA-QDA). By using the leaves, the F-score values was slightly lower (0.79) applying PCA and linear discriminant analysis (PCA-LDA). It was possible to use NIR spectroscopy associated with multivariate techniques as a non-destructive method to determine the sex types in papaya trees using both seeds and leaves of the seedlings.
Descrição
Palavras-chave
Carica papaya L, Chemometrics, PCA-LDA, PCA-QDA, Sexing
Como citar
Computers and Electronics in Agriculture, v. 193.