Application of hyperspectral imaging for nondestructive measurement of plum quality attributes
By Bo Li, Magdalena Cobo-Medina, J. Lecourt, N. Harrison, R. Harrison, J. Cross. Postharvest Biology and Technology, 2018
70 citations recorded by the source corpus.
Abstract Colour, firmness and soluble solid content (SSC) are three important quality attributes of fruit that affect consumer acceptance. However, measurement of these attributes largely relies on destructive manual assessment, which is time consuming, and can only be applied to small number of batches. In this study, two hyperspectral cameras in the visible and near infrared (VNIR) regions between 600–975 nm and the short wave near infrared (SWIR) region between 865–1610 nm were evaluated for the non-destructive quantification of colour (L*, a* and b*), firmness and SSC. In total, images of 354 ‘Victoria’ and ‘Marjorie’s Seedling’ plums were collected for the calibration and validation of partial least square regression (PLSR) models. The performance of the prediction models was compared for both the cultivars alone and in combination. The effect of a light scattering correction on spherical objects was also investigated. This study showed that the SWIR hyperspectral imaging could accurately predict SSC with correlation coefficients for prediction ( r p 2 ) greater than 0.8, while VNIR hyperspectral imaging showed a better correlation with colour with r p 2 values greater than 0.7 for L* and a*. This study shows that the use of hyperspectral imaging is feasible to non-destructively predict the SSC and the colour of two plums cultivars with high accuracy.
Subfields: Analytical Chemistry, Ecology, Industrial and Manufacturing Engineering
View Paper: https://www.semanticscholar.org/paper/6164a33014d9ad8de46a552d59c12eafd82cffc0
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Analytical Chemistry
In Industrial and Manufacturing Engineering
Citations in the corpus
- Machine learning-based image processing for on-line defect recognition in additive manufacturing (2019, background): “For the time being, researchers are still working on identifying and classifying different kinds of product flaws [5].”
- Using Deep Learning to Detect Defects in Manufacturing: A Comprehensive Survey and Current Challenges (2020, background): “summarized the application of hyperspectral [9], vibration spectrum [10], infrared [11], and other spectral technologies.”