Determination of soluble solids content in Prunus avium by Vis/NIR equipment using linear and non-linear regression methods

• Main Authors: Victoria Lafuente, Luis J. Herrera, Jesús Val, Razvan Ghinea, Angel I. Negueruela
• Format: Article
• Language: English
• Published: Instituto Nacional de Investigación y Tecnología Agraria y Alimentaria 2020-02-01
• Series: Spanish Journal of Agricultural Research
• Subjects: pls; ls-svm; selection variables
• Online Access: http://revistas.inia.es/index.php/sjar/article/view/13891
• ISSN: 2171-9292

Aim of study: Developing models to determine soluble solids content (SSC) in cherry trees by means of Vis/NIR spectroscopy. Area of study: The Spanish Autonomous Community of Aragón (Spain). Material and methods: Vis/NIR spectroscopy was applied to Prunus avium fruit ‘Chelan’ (n=360) to predict total SSC using a range 400-2420 nm. Linear (PLS) and nonlinear (LSSVM) regression methods were applied to establish prediction models. Main results: The two regression methods applied obtained similar results (Rcv2=0.97 and Rcv2=0.98 respectively). The range 700-1060 nm attained better results to predict SSC in different seasons. Forty variables selected according to the variable selection method achieved Rcv2 value, 0.97 similar than full range. Research highlights: The development of this methodology is of great interest to the fruit sector in the area, facilitating the harvest for future seasons. Further work is needed on the development of the NIRS methodology and on new calibration equations for other varieties of cherry and other species.