Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging
Protein content is one of the most crucial factors in soybean quality. However, the breeding procedure necessitates the time-consuming and costly selection of elite genotypes from many experimental lines in a destructive manner. The present work aims to predict protein content in single soybean seed...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier B.V.
2022
|
| Subjects: | |
| Online Access: | https://repository.ugm.ac.id/282850/1/56_Non%20destructive%20prediction.pdf |
| _version_ | 1797037609433694208 |
|---|---|
| author | Aulia, Rizkiana Kim, Yena Zuhrotul Amanah, Hanim Muhammad Akbar Andi, Arief Kim, Haeun Kim, Hangi Lee, Wang-Hee Kim, Kyung-Hwan Baek, Jeong-Ho Cho, Byoung-Kwan |
| author_facet | Aulia, Rizkiana Kim, Yena Zuhrotul Amanah, Hanim Muhammad Akbar Andi, Arief Kim, Haeun Kim, Hangi Lee, Wang-Hee Kim, Kyung-Hwan Baek, Jeong-Ho Cho, Byoung-Kwan |
| author_sort | Aulia, Rizkiana |
| collection | UGM |
| description | Protein content is one of the most crucial factors in soybean quality. However, the breeding procedure necessitates the time-consuming and costly selection of elite genotypes from many experimental lines in a destructive manner. The present work aims to predict protein content in single soybean seeds non-destructively using Near-Infrared (NIR) Hyperspectral Imaging (HSI). 1491 seed samples from 3 varieties of the low, medium, and high protein content (consisting of 371, 560, and 560 samples, respectively) were measured using the NIR-HSI system with a range of 900–1800 nm. The spectral data extracted from the HSI 3D hypercube were synchronised to the reference values examined from chemical analysis. The calibration model was constructed using partial least square regression (PLSR) methods based on the 70% spectral data and then validated using the remaining 30% of data. The result showed that the NIR-HSI technique is a promising method to predict protein content in soybean seeds, as shown by an R2 of 0.92 and a root mean square error (RMSE) of 1.08% . In addition, the chemical images visualised the distribution of protein content for the multiple soybean seed showed the possibility of the developed technique for the use of rapid evaluation of massive samples in the processing line. |
| first_indexed | 2024-03-14T00:05:56Z |
| format | Article |
| id | oai:generic.eprints.org:282850 |
| institution | Universiti Gadjah Mada |
| language | English |
| last_indexed | 2024-03-14T00:05:56Z |
| publishDate | 2022 |
| publisher | Elsevier B.V. |
| record_format | dspace |
| spelling | oai:generic.eprints.org:2828502023-11-17T01:17:18Z https://repository.ugm.ac.id/282850/ Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging Aulia, Rizkiana Kim, Yena Zuhrotul Amanah, Hanim Muhammad Akbar Andi, Arief Kim, Haeun Kim, Hangi Lee, Wang-Hee Kim, Kyung-Hwan Baek, Jeong-Ho Cho, Byoung-Kwan Food technology Protein content is one of the most crucial factors in soybean quality. However, the breeding procedure necessitates the time-consuming and costly selection of elite genotypes from many experimental lines in a destructive manner. The present work aims to predict protein content in single soybean seeds non-destructively using Near-Infrared (NIR) Hyperspectral Imaging (HSI). 1491 seed samples from 3 varieties of the low, medium, and high protein content (consisting of 371, 560, and 560 samples, respectively) were measured using the NIR-HSI system with a range of 900–1800 nm. The spectral data extracted from the HSI 3D hypercube were synchronised to the reference values examined from chemical analysis. The calibration model was constructed using partial least square regression (PLSR) methods based on the 70% spectral data and then validated using the remaining 30% of data. The result showed that the NIR-HSI technique is a promising method to predict protein content in soybean seeds, as shown by an R2 of 0.92 and a root mean square error (RMSE) of 1.08% . In addition, the chemical images visualised the distribution of protein content for the multiple soybean seed showed the possibility of the developed technique for the use of rapid evaluation of massive samples in the processing line. Elsevier B.V. 2022-12 Article PeerReviewed application/pdf en https://repository.ugm.ac.id/282850/1/56_Non%20destructive%20prediction.pdf Aulia, Rizkiana and Kim, Yena and Zuhrotul Amanah, Hanim and Muhammad Akbar Andi, Arief and Kim, Haeun and Kim, Hangi and Lee, Wang-Hee and Kim, Kyung-Hwan and Baek, Jeong-Ho and Cho, Byoung-Kwan (2022) Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging. Infrared Physics and Technology, 127. pp. 1-11. ISSN 3504495 DOI 10.1016/j.infrared.2022.104365 |
| spellingShingle | Food technology Aulia, Rizkiana Kim, Yena Zuhrotul Amanah, Hanim Muhammad Akbar Andi, Arief Kim, Haeun Kim, Hangi Lee, Wang-Hee Kim, Kyung-Hwan Baek, Jeong-Ho Cho, Byoung-Kwan Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging |
| title | Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging |
| title_full | Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging |
| title_fullStr | Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging |
| title_full_unstemmed | Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging |
| title_short | Non-destructive prediction of protein contents of soybean seeds using near-infrared hyperspectral imaging |
| title_sort | non destructive prediction of protein contents of soybean seeds using near infrared hyperspectral imaging |
| topic | Food technology |
| url | https://repository.ugm.ac.id/282850/1/56_Non%20destructive%20prediction.pdf |
| work_keys_str_mv | AT auliarizkiana nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT kimyena nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT zuhrotulamanahhanim nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT muhammadakbarandiarief nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT kimhaeun nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT kimhangi nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT leewanghee nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT kimkyunghwan nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT baekjeongho nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging AT chobyoungkwan nondestructivepredictionofproteincontentsofsoybeanseedsusingnearinfraredhyperspectralimaging |