Application of calibrations to hyperspectral images of food grains: example for wheat falling number
The presence of a few kernels with sprouting problems in a batch of wheat can result in enzymatic activity sufficient to compromise flour functionality and bread quality. This is commonly assessed using the Hagberg Falling Number (HFN) method, which is a batch analysis. Hyperspectral imaging (HSI) c...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Published: |
IM Publications
2017
|
| Subjects: | |
| Online Access: | https://eprints.nottingham.ac.uk/42284/ |
| _version_ | 1848796453644795904 |
|---|---|
| author | Caporaso, Nicola Whitworth, Martin B. Fisk, Ian D. |
| author_facet | Caporaso, Nicola Whitworth, Martin B. Fisk, Ian D. |
| author_sort | Caporaso, Nicola |
| building | Nottingham Research Data Repository |
| collection | Online Access |
| description | The presence of a few kernels with sprouting problems in a batch of wheat can result in enzymatic activity sufficient to compromise flour functionality and bread quality. This is commonly assessed using the Hagberg Falling Number (HFN) method, which is a batch analysis. Hyperspectral imaging (HSI) can provide analysis at the single grain level with potential for improved performance. The present paper deals with the development and application of calibrations obtained using an HSI system working in the near infrared (NIR) region (~900–2500 nm) and reference measurements of HFN. A partial least squares regression calibration has been built using 425 wheat samples with a HFN range of 62–318 s, including field and laboratory pre-germinated samples placed under wet conditions. Two different approaches were tested to apply calibrations: i) application of the calibration to each pixel, followed by calculation of the average of the resulting values for each object (kernel); ii) calculation of the average spectrum for each object, followed by application of the calibration to the mean spectrum. The calibration performance achieved for HFN (R2 = 0.6; RMSEC ~ 50 s; RMSEP ~ 63 s) compares favourably with other studies using NIR spectroscopy. Linear spectral pre-treatments lead to similar results when applying the two methods, while non-linear treatments such as standard normal variant showed obvious differences between these approaches. A classification model based on linear discriminant analysis (LDA) was also applied to segregate wheat kernels into low (<250 s) and high (>250 s) HFN groups. LDA correctly classified 86.4% of the samples, with a classification accuracy of 97.9% when using HFN threshold of 150 s. These results are promising in terms of wheat quality assessment using a rapid and non-destructive technique which is able to analyse wheat properties on a single-kernel basis, and to classify samples as acceptable or unacceptable for flour production. |
| first_indexed | 2025-11-14T19:48:14Z |
| format | Article |
| id | nottingham-42284 |
| institution | University of Nottingham Malaysia Campus |
| institution_category | Local University |
| last_indexed | 2025-11-14T19:48:14Z |
| publishDate | 2017 |
| publisher | IM Publications |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | nottingham-422842024-08-15T15:22:29Z https://eprints.nottingham.ac.uk/42284/ Application of calibrations to hyperspectral images of food grains: example for wheat falling number Caporaso, Nicola Whitworth, Martin B. Fisk, Ian D. The presence of a few kernels with sprouting problems in a batch of wheat can result in enzymatic activity sufficient to compromise flour functionality and bread quality. This is commonly assessed using the Hagberg Falling Number (HFN) method, which is a batch analysis. Hyperspectral imaging (HSI) can provide analysis at the single grain level with potential for improved performance. The present paper deals with the development and application of calibrations obtained using an HSI system working in the near infrared (NIR) region (~900–2500 nm) and reference measurements of HFN. A partial least squares regression calibration has been built using 425 wheat samples with a HFN range of 62–318 s, including field and laboratory pre-germinated samples placed under wet conditions. Two different approaches were tested to apply calibrations: i) application of the calibration to each pixel, followed by calculation of the average of the resulting values for each object (kernel); ii) calculation of the average spectrum for each object, followed by application of the calibration to the mean spectrum. The calibration performance achieved for HFN (R2 = 0.6; RMSEC ~ 50 s; RMSEP ~ 63 s) compares favourably with other studies using NIR spectroscopy. Linear spectral pre-treatments lead to similar results when applying the two methods, while non-linear treatments such as standard normal variant showed obvious differences between these approaches. A classification model based on linear discriminant analysis (LDA) was also applied to segregate wheat kernels into low (<250 s) and high (>250 s) HFN groups. LDA correctly classified 86.4% of the samples, with a classification accuracy of 97.9% when using HFN threshold of 150 s. These results are promising in terms of wheat quality assessment using a rapid and non-destructive technique which is able to analyse wheat properties on a single-kernel basis, and to classify samples as acceptable or unacceptable for flour production. IM Publications 2017-04-28 Article PeerReviewed Caporaso, Nicola, Whitworth, Martin B. and Fisk, Ian D. (2017) Application of calibrations to hyperspectral images of food grains: example for wheat falling number. Journal of Spectral Imaging, 6 (a4). pp. 1-15. ISSN 2040-4565 Hyperspectral imaging Partial least squares calibration Food grains Whole wheat kernels Hagberg Falling Number (HFN) https://www.impublications.com/mjsi/jsi-abstract.php?code=I06_a4 doi:10.1255/jsi.2017.a4 doi:10.1255/jsi.2017.a4 |
| spellingShingle | Hyperspectral imaging Partial least squares calibration Food grains Whole wheat kernels Hagberg Falling Number (HFN) Caporaso, Nicola Whitworth, Martin B. Fisk, Ian D. Application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| title | Application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| title_full | Application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| title_fullStr | Application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| title_full_unstemmed | Application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| title_short | Application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| title_sort | application of calibrations to hyperspectral images of food grains: example for wheat falling number |
| topic | Hyperspectral imaging Partial least squares calibration Food grains Whole wheat kernels Hagberg Falling Number (HFN) |
| url | https://eprints.nottingham.ac.uk/42284/ https://eprints.nottingham.ac.uk/42284/ https://eprints.nottingham.ac.uk/42284/ |