FT-NIR and linear discriminant analysis to classify chickpea seeds produced with harvest aid chemicals

• Published in: Food chemistry Vol. 342; p. 128324
• Main Authors: Ribeiro, João Paulo Oliveira, Medeiros, André Dantas de, Caliari, Italo Pelição, Trancoso, Ana Clara Reis, Miranda, Rafaela Marques de, Freitas, Francisco Claudio Lopes de, Silva, Laércio Junio da, Dias, Denise Cunha Fernandes dos Santos
• Format: Journal Article
• Language: English
• Published: England Elsevier Ltd 16-04-2021
• Subjects: Algorithms; Cicer - embryology; Discriminant Analysis; Edible Grain; Fourier Analysis; Near-infrared spectroscopy; Pesticide residues; Pre-harvest chickpeas; Seed physiological changes; Seeds - chemistry; Seeds - classification; Spectroscopy, Near-Infrared - methods; Systemic and contact herbicides; Systemic and contact herbicides; Pre-harvest chickpeas; Pesticide residues; Seed physiological changes; Near-infrared spectroscopy
• ISSN: 0308-8146; 1873-7072
• DOI: 10.1016/j.foodchem.2020.128324

[Display omitted]
•FT-NIR data were used to classify chickpea seeds with changes caused by herbicides.•Spectral-based prediction of chickpea seed classes reached 94% accuracy.•The desiccant type and phenological stage of application impact on the seed physiological quality.•The proposed approach assesses seed properties quick and non-destructive manner.
Spectroscopy and machine learning (ML) algorithms have provided significant advances to the modern food industry. Instruments focusing on near-infrared spectroscopy allow obtaining information about seed and grain chemical composition, which can be related to changes caused by field pesticides. We investigated the potential of FT-NIR spectroscopy combined with Linear Discriminant Analysis (LDA) to discriminate chickpea seeds produced using different desiccant herbicides at harvest anticipation. Five herbicides applied at three moments of the plant reproductive stage were utilized. The NIR spectra obtained from individual seeds were used to build ML models based on LDA algorithm. The models developed to identify the herbicide and the plant phenological stage at which it was applied reached 94% in the independent validation set. Thus, the LDA models developed using near-infrared spectral data provided to be efficient, quick, non-destructive, and accurate to identify differences between seeds due to pre-harvest herbicides application.