Near-infrared hyperspectral imaging combined with machine learning for physicochemical-based quality evaluation of durian pulp

Simple item page
dc.contributor.authorSharma, Sneha
dc.contributor.authorSirisomboon, Panmanas
dc.contributor.authorK.C, Sumesh
dc.contributor.authorTerdwongworakul, Anupun
dc.contributor.authorPhetpan, Kittisak
dc.contributor.authorKshetri, Tek Bahadur
dc.contributor.authorSangwanangkul, Peerapong
dc.date.accessioned2026-08-06T10:41:42Z
dc.date.available2026-08-06T10:41:42Z
dc.date.issued2023-06-01
dc.description.abstractThis research reports on the application of near-infrared hyperspectral imaging (NIR-HSI) system for predicting the physicochemical properties; dry matter (DM), total soluble solids (TSS), and fat content (FC) of durian. Partial least squares regression (PLSR), support vector machine (SVM), random forest (RF), and 1D convolution neural network (CNN) models: custom, U-Net, and VGG19; were developed to predict DM, TSS, and FC of durian pulp. Feature wavelengths were selected using a genetic algorithm (GA) and successive projection algorithm (SPA). The selected wavelengths were then validated based on the algorithms for regression model development. GA-PLSR model was compelling to predict the DM and FC in durian pulp, which obtained the coefficient of determination for the test set (r<sup>2</sup>) and root mean square error of prediction (RMSEP) of 0.97 and 1.12% for DM and 0.86 and 0.64% for FC, respectively. The GA-PLSR model provided the best result for the TSS prediction with r<sup>2</sup>, and RMSEP of 0.90 and 1.40%, respectively, whereas the SPA-PLSR model based on only thirteen wavelengths attained fair result with the r<sup>2</sup> and RMSEP of 0.79 and 2.03%, respectively. The above results show that the pushbroom NIR-HSI system achieved promising results for estimating DM, TSS, and FC in durian pulp. This research identified the featured wavelengths that can be used to develop a portable and reliable HSI or multispectral system to be installed at durian packaging firms for quality inspection and grading.
dc.identifier.citationPostharvest Biology and Technology, 200, 2023
dc.identifier.doi10.1016/j.postharvbio.2023.112334
dc.identifier.issn09255214
dc.identifier.other2-s2.0-85151476819
dc.identifier.urihttps://dspace.kmitl.ac.th/handle/123456789/14475
dc.sourcePostharvest Biology and Technology
dc.subjectDurian
dc.subjectHyperspectral imaging
dc.subjectMachine learning
dc.subjectNear-infrared
dc.subjectPhysicochemical properties
dc.titleNear-infrared hyperspectral imaging combined with machine learning for physicochemical-based quality evaluation of durian pulp
dc.typeArticle

Files

Collections

Research