Indirect prediction of pulp dry matter content (DM) was investigated using near-infrared spectra acquired from fruit stems. The spectral difference between stem and pulp spectra, together with pulp DM, was used to adjust the stem spectra, thereby integrating pulp-related information into the stem spectral data. Partial least squares regression models were developed using unprocessed, preprocessed, and adjusted stem spectra for comparison. The adjusted spectra model achieved a prediction correlation coefficient of 0.828 and a root mean square error of prediction of 1.630%, marginally outperforming the unprocessed and preprocessed models. The best preprocessed model and the adjusted spectra model were subsequently used to build support vector machine classifiers. In binary classification, the adjusted spectra model showed slightly improved performance on the test set, with average accuracy, precision, and recall values of 0.882, 0.833, and 0.923, respectively. This study should be viewed as a proof of concept. Future validation using independent samples from different seasons, orchards, production lots, and environmental conditions is necessary to confirm the robustness and generalizability of the approach.
Citation: Kaewkarn Phuangsombut, Arthit Phuangsombut, Sirinad Noypitak, Anupun Terdwongworakul. Spectral difference-based adjustment of stem near-infrared spectra for indirect prediction of durian pulp dry matter and maturity[J]. AIMS Agriculture and Food, 2026, 11(3): 510-530. doi: 10.3934/agrfood.2026026
Indirect prediction of pulp dry matter content (DM) was investigated using near-infrared spectra acquired from fruit stems. The spectral difference between stem and pulp spectra, together with pulp DM, was used to adjust the stem spectra, thereby integrating pulp-related information into the stem spectral data. Partial least squares regression models were developed using unprocessed, preprocessed, and adjusted stem spectra for comparison. The adjusted spectra model achieved a prediction correlation coefficient of 0.828 and a root mean square error of prediction of 1.630%, marginally outperforming the unprocessed and preprocessed models. The best preprocessed model and the adjusted spectra model were subsequently used to build support vector machine classifiers. In binary classification, the adjusted spectra model showed slightly improved performance on the test set, with average accuracy, precision, and recall values of 0.882, 0.833, and 0.923, respectively. This study should be viewed as a proof of concept. Future validation using independent samples from different seasons, orchards, production lots, and environmental conditions is necessary to confirm the robustness and generalizability of the approach.
| [1] | Maninang JS, Wongs-Aree C, Kanlayanarat S, et al. (2011) Influence of maturity and postharvest treatment on the volatile profile and physiological properties of the durian fruit. Int Food Res J 18: 1067–1075. |
| [2] | Department of Agricultural Extension (DOAE) (2026) Summary of the fruit production season situation report. Fruit Situation Reporting System. Available from: https://simplefruit.doae.go.th/dashboard/index. |
| [3] | Percival SS, Findley B (2007) What's in Your Tropical Fruit? FSHN 07-08/FS144, 9/2007. EDIS 2007: No. 19.https://doi.org/10.32473/edis-fs144-2007 |
| [4] | Pascua OC, Cantila MS (1992) Maturity indices of durian (durio zibethinus murray). Philipp J Crop Sci 17: 119–124. |
| [5] | Siriphanich J (2011) Durian (Durio zibethinus Merr.). In: Yahia EM (Ed.), Postharvest biology and technology of tropical and subtropical fruits, Woodhead Publishing, 80–114.https://doi.org/10.1533/9780857092885.80 |
| [6] | National Bureau of Agricultural Commodity and Food Standards (2003) Thai Agricultural Standard TAS 3-2003: Durian. R Thai Gov Gaz 120(115D). |
| [7] |
Onsawai P, Phetpan K, Khurnpoon L, et al. (2021) Evaluation of physiological properties and texture traits of durian pulp using near-infrared spectra of the pulp and intact fruit. Measurement 174: 108684.https://doi.org/10.1016/j.measurement.2020.108684 doi: 10.1016/j.measurement.2020.108684
|
| [8] |
Pokhrel DR, Sirisomboon P, Khurnpoon L, et al. (2023) Comparing machine learning and PLSDA algorithms for durian pulp classification using inline NIR spectra. Sensors 23: 5327.https://doi.org/10.3390/s23115327 doi: 10.3390/s23115327
|
| [9] |
Puttipipatkajorn A, Terdwongworakul A, Puttipipatkajorn A, et al. (2023) Indirect prediction of dry matter in durian pulp with combined features using miniature NIR spectrophotometer. IEEE Access 11: 84810–84821.https://doi.org/10.1109/ACCESS.2023.3303020 doi: 10.1109/ACCESS.2023.3303020
|
| [10] |
Saenphon C, Ditcharoen S, Malai C, et al. (2023) Total soluble solids, dry matter content prediction and maturity stage classification of durian fruit using long-wavelength NIR reflectance. J Food Compos Anal 124: 105667.https://doi.org/10.1016/j.jfca.2023.105667 doi: 10.1016/j.jfca.2023.105667
|
| [11] |
Posom J, Saenphon C, Ditcharoen S, et al. (2025) Deep neural networks (DNNs) chemical compositions estimation of fresh durian in-line via near infrared spectroscopy. J Food Compos Anal 142: 107410.https://doi.org/10.1016/j.jfca.2025.107410 doi: 10.1016/j.jfca.2025.107410
|
| [12] |
Ditcharoen S, Sirisomboon P, Saengprachatanarug K, et al. (2023) Improving the non-destructive maturity classification model for durian fruit using near-infrared spectroscopy. Artif Intell Agric 7: 35–43.https://doi.org/10.1016/j.aiia.2023.02.002 doi: 10.1016/j.aiia.2023.02.002
|
| [13] |
Imai M, Sangsoy K, Blasco J, et al. (2025) Enhancing dry matter prediction in durian using peduncle-based NIR spectroscopy across maturation rates. Postharvest Biol Technol 230: 113773.https://doi.org/10.1016/j.postharvbio.2025.113773 doi: 10.1016/j.postharvbio.2025.113773
|
| [14] | Sangwanangkul P, Siriphanich J (2000) Growth and maturation of durian fruit cv. Monthong. Thai J Agric Sci 33: 75–82. |
| [15] | Meier KJ, Brudney JL, Bohte J (2009) Applied statistics for public and nonprofit administration. Cengage Learning, Wadsworth, Boston. |
| [16] | Goodwin CJ, Goodwin KA (2020) Research in psychology: methods and design. 9th ed., Wiley. |
| [17] |
Yongyut N, Baopa P, Meetha S, et al. (2025) Fruit quality and antioxidant content in durian (Durio zibethinus Murr.) cv. 'Monthong' in different maturity stages. Horticulturae 11: 432.https://doi.org/10.3390/horticulturae11040432 doi: 10.3390/horticulturae11040432
|
| [18] | Ali MM, Hashim N, Shahamshah MI (2021) Durian (Durio zibethinus) ripeness detection using thermal imaging with multivariate analysis. Postharvest Biol Technol 176: 111517.https://doi.org/10.1016/j.postharvbio.2021.111517 |
| [19] | Smola AJ, Schölkopf B (2004) A tutorial on support vector regression. Stat Comput 14: 199–222.https://doi.org/10.1023/B: ASTCO.0000035301.49549.88 |
| [20] | Raschka S, Mirjalili V (2019) Python machine learning: Machine learning and deep learning with Python, scikit-learn, and TensorFlow 2. Packt Publishing Ltd., Birmingham. |
| [21] | Workman J, Weyer L (2007) Practical guide to interpretive near-infrared spectroscopy. CRC Press. |
| [22] | Ciurczak EW, Igne B, Workman J Jr, et al. (Eds.) (2021) Handbook of near-infrared analysis. CRC Press, Boca Raton, FL. |
| [23] | Tsuchikawa S, Siesler HW (2006) Near-infrared spectroscopic monitoring of the diffusion process of deuterium-labeled molecules in wood. In: Stokke DD, Groom LH (Eds.), Characterization of the Cellulosic Cell Wall, Blackwell Publishing, 123–137.https://doi.org/10.1002/9780470999714.ch10 |
| [24] | Chattavongsin R, Siriphanich J (1987) Fruit-stem anatomy of durians at various harvesting stages [in Thai with English abstract]. In: Proceedings of the 26th Academic Conference, Kasetsart University, Bangkok, 405–412. |
| [25] | Beck CB (2010) An introduction to plant structure and development: Plant anatomy for the twenty-first century. Cambridge University Press, Cambridge, UK, 17–24.https://doi.org/10.1017/CBO9780511844683 |
| [26] | Williams PC, Norris KH (2001) Near-infrared technology in the agricultural and food industries. 2nd ed., American Association of Cereal Chemists. |
| [27] | Wold S, Johansson E, Cocchi M (1993) PLS-partial least squares projections to latent structures. In: 3D QSAR in Drug Design: Theory, Methods and Applications, Elsevier. |
| [28] |
Currà A, Gasbarrone R, Bonifazi G, et al. (2022) Near-infrared transflectance spectroscopy discriminates solutions containing two commercial formulations of botulinum toxin type A diluted at recommended volumes for clinical reconstitution. Biosensors 12: 216.https://doi.org/10.3390/bios12040216 doi: 10.3390/bios12040216
|
| [29] | Beć KB, Grabska J, Huck CW, (2020) Near-infrared spectroscopy in bio-applications. Molecules 25(12): 2948.https://doi.org/10.3390/molecules25122948 |
| [30] | Osborne B, Fearn T, Hindle PH (1993) Practical NIR spectroscopy with applications in food and beverage analysis, 2nd ed., Addison-Wesley Longman Ltd. : Harlow, UK, 227. |
| [31] | Sokolova M, Lapalme G (2009) A systematic analysis of performance measures for classification tasks. Inf Process Manag 45: 427–437.https://doi.org/10.1016/j.ipm.2009.03.002 |
Figures(8) / Tables(6)
Kaewkarn Phuangsombut, Arthit Phuangsombut, Sirinad Noypitak, Anupun Terdwongworakul. Spectral difference-based adjustment of stem near-infrared spectra for indirect prediction of durian pulp dry matter and maturity[J]. AIMS Agriculture and Food, 2026, 11(3): 510-530. doi: 10.3934/agrfood.2026026
DownLoad: