Non-invasive spectroscopic methods to estimate orange firmness, peel thickness, and total pectin content
Orange firmness, peel thickness, and total pectin content are associated with fruit quality and denote important parameters for the food industry. These attributes are usually determined through destructive methods that can be time-consuming and also unable to monitor fruit quality over time. Theref...
| Autores: | , , , |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2017 |
| País: | Brasil |
| Institución: | Universidade Estadual Paulista (UNESP) |
| Repositorio: | Repositório Institucional da UNESP |
| Idioma: | inglés |
| OAI Identifier: | oai:repositorio.unesp.br:11449/174385 |
| Acceso en línea: | http://dx.doi.org/10.1016/j.microc.2017.03.039 http://hdl.handle.net/11449/174385 |
| Access Level: | acceso abierto |
| Palabra clave: | MIR NIR Orange PLSR Quality TD-NMR |
| Sumario: | Orange firmness, peel thickness, and total pectin content are associated with fruit quality and denote important parameters for the food industry. These attributes are usually determined through destructive methods that can be time-consuming and also unable to monitor fruit quality over time. Therefore, non-invasive methods such time-domain nuclear magnetic resonance (TD-NMR), near-infrared (NIR), and mid-infrared (MIR) spectroscopies may represent efficient alternatives to evaluate these quality attributes. In this work, partial least square regression (PLSR) models of TD-NMR relaxometry as well as NIR and MIR spectroscopic data were used to predict firmness, peel thickness, and total pectin content of fresh Valencia oranges. Principal component analyses (PCA) were applied to explain the correlations of orange ripening stage, flowering, and crop season with its physicochemical parameters. Data obtained through standard destructive methods were used to calibrate and validate the PLSR models. NIR and MIR showed the best PLSR models for orange firmness, with Pearson correlation coefficients (r) of 0.92 and 0.84 and squared errors of prediction (SEP) equal to 6.22 and 9.05 N, respectively. Orange peel thickness PLSR model was validated only by TD-NMR (r = 0.72; SEP = 0.49 mm). TD-NMR and NIR also presented potential to predict total pectin orange in orange (r = 0.76 and 0.70; SEP = 5.76% and 5.04%, respectively). Therefore, NIR presented a higher potential to predict orange firmness than MIR and TD-NMR. On the other hand, TD-NMR showed a higher prediction power concerning peel thickness than NIR and MIR. Both NIR and TD-NMR methods showed similar prediction powers for total pectin content. |
|---|