KMITL

Permanent URI for this communityhttps://dspace.kmitl.ac.th/handle/123456789/1

Browse

Search Results

Now showing 1 - 8 of 8

Some of the metrics are blocked by your
consent settings
Investigation of physiological disorder classification in mangosteen fruit using visible and shortwave near-infrared spectroscopy combined with machine learning
(2025-12-01)
Ruttanadech, Nuttapong
;
Momin, Abdul
;
Phetpan, Kittisak
;
Chaichanyut, Montree
;
Thongphut, Chitwadee
Accurate classification of physiological disorders in mangosteen fruit is crucial for ensuring production quality, safety, sustainability, and economic viability. This study investigates the application of visible and shortwave near-infrared (Vis/SWNIR) reflectance spectroscopy, combined with machine learning algorithms, to classify three primary disorders: normal fruit (NF), translucent flesh disorder (TFD), and TFD with yellow gummy latex (TFD & YGL). The study specifically examines the effects of light intensity, spectral pretreatments, and machine learning models on classification performance. Spectral data were collected using two light intensities (50 % and 100 % of a 150 W light source) and processed with three pretreatments: standard normal variate (SNV), second derivative Savitzky-Golay (SGD2), and a combination of SNV and SGD2. Random forest (RF), support vector machine (SVM), and multi-layer perceptron (MLP) algorithms were used for classification. The SGD2 method improved differentiation, especially for the TFD & YGL class, in the 700–725 nm wavelength range, which is associated with xanthone content in the fruit's pericarp. Higher light intensity (100 %) significantly improved classification accuracy, achieving an overall accuracy of 0.71 and an average F1 score of 0.61 with the RF model. Despite these improvements, the model struggled to distinguish the TFD class from NF due to their similar spectral profiles. Overall, the Vis/SWNIR spectroscopy and machine learning combination shows strong potential for the non-destructive classification of mangosteen fruit disorders. Both light intensity and spectral pretreatments play critical roles in enhancing performance. Future studies should focus on improving spectral sensitivity to better capture internal fruit characteristics.
Some of the metrics are blocked by your
consent settings
Near-infrared hyperspectral imaging for predicting the quality of SO2 pre-treated and dehydrated mango
(2025-08-01)
Aozora, Wayan Dipasasri
;
Tantinantrakun, Achiraya
;
Thompson, Anthony Keith
;
Teerachaichayut, Sontisuk
Prediction for quality indices of SO<inf>2</inf> pre-treated and dehydrated mango was accessed by NIR-HSI. Models for predicting TSS and SO<inf>2</inf> content achieved R = 0.82; RMSEP = 2.42% and R = 0.83; RMSEP = 56.40 mg/kg, respectively. Visualization of TSS and SO<inf>2</inf> content could be presented by predictive images.
Some of the metrics are blocked by your
consent settings
Prediction of water activity in mamón (Filipino sponge) cakes by near infrared hyperspectral imaging
(2020-01-01)
Sricharoonratana, Manunchaya
;
Teerachaichayut, Sontisuk
Water activity in foods can result in detrimental microbial activity during storage. The usual methods of water activity measurement involve destruction of the sample. Near infrared (NIR) hyperspectral imaging has previously been successfully used as a non-destructive method to determine various physical and chemical characteristics of a variety of foods. Therefore, this method was tested to determine whether it could be used to measure water activity of mamón cakes, a popular sponge cake developed in the Philippines. Individual samples (n = 178) were divided into a calibration set (n=119) and a prediction set (n=59). These samples were tested using NIR hyperspectral imaging (935-1720 nm) with a smoothing spectral pretreatment selected for developing the calibration model. Partial least squares regression was used to establish the model in order to predict the water activity. The results showed the accuracy of the calibration model in prediction that gave a correlation coefficient of 0.767 and the root mean square error of prediction of 0.0130. It was therefore concluded that NIR hyperspectral imaging has a potential for use and application for measuring the water activity of mamón cakes.
Some of the metrics are blocked by your
consent settings
Non-destructive prediction of pH and total soluble solids of lime [Citrus × aurantifolia (Cristm.) Swinge] by visible and near-infrared spectroscopy
(2017-11-25)
Huong, H. T.
;
Teerachaichayut, S.
The non-destructive visible and near-infrared spectroscopy (Vis/NIRS) technique is well suited for evaluating various internal quality indices of fruits quickly and accurately. The objective of this study was to evaluate the relationships between Vis/NIR measurements and the internal quality indices of lime, including pH and total soluble solids (TSS, °Brix). For this experiment, reflectance measurement in the 400-2500 nm range was done on 140 samples for pH and 117 samples for TSS. Partial least square (PLS) regression was used to establish the calibration models. First-order derivative and multiplicative scatter correction spectral pretreatments were used to develop calibration models for pH and TSS, respectively. The correlation coefficient (R) and the root mean square error of prediction (RMSEP) from the calibration model for pH were 0.95 and 0.06. The corresponding values for TSS were 0.81 and 0.24 °Brix, respectively. The results showed that Vis/NIRS measurements in the spectral range 400-2500 nm could be used to access pH and TSS of lime.
Some of the metrics are blocked by your
consent settings
Grouping marian plums harvested at different times by transmittance near-infrared spectroscopy
(2017-11-25)
Teerachaichayut, S.
;
Phonmakham, S.
;
Suktanarak, S.
Marian plum (Bouea burmanica Griff.) ‘Toon Klaow’ is one of Thailand’s favorite fruits. Marian plum’s edible quality depends strongly on its harvest time. This study investigated a non-destructive technique for classifying marian plums according to their harvest time after the day that their blossom set. The non-destructive technique used was transmittance mode, short wavelength near-infrared (SW-NIR) spectroscopy in the wavelength range 660-960 nm. Marian plum samples (n=110) were harvested at 62, 65, 68 and 73 days after flowering. The soluble solids content (SSC) and titratable acid (TA) were determined accurately by standard methods. SW-NIR spectra of these samples were obtained and analyzed by principle component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). The following spectral pretreatments were applied, standard normal variate (SNV), smoothing (Savitsky-Golay), and first derivative, in order to obtain optimal grouping results. The PC1 and PC2 score plot of the PCA could not clearly separate some of the classified groups. For the results of PLS-DA, its cross-validated grouping accuracy was R=0.91 and RMSECV=1.28; hence, it can be concluded that SW-NIR spectroscopy has good potential for determining the harvest time of marian plums.
Some of the metrics are blocked by your
consent settings
Detection of internal mold infection in tomato by transmittance near infrared spectroscopy
(2014-10-20)
Jannok, P.
;
Petcharaporn, K.
;
Teerachaichayut, S.
Alternaria alternata is the black mold occurring inside tomato. This defect can be normally found by destructive method but it cannot be detected by visible inspection from outside appearance of intact tomato. Therefore, a non-destructive technique for prediction of internal mold infection in tomato is required. Near infrared (NIR) spectroscopy technique was considered in this research. Transmittance NIR spectra in the range of 665-955 nm of tomato were acquired. Partial least squares-discriminant analysis (PLS-DA) was performed to establish the calibration model. Results indicated that combination of the standard normal variate transformation (SNV) and smoothing (Savitzky-Golay) pretreatment appeared the best method to develop the model. The calibration model was crossvalidated by a training set (N=140) and used for prediction by a test set (N=60). It obtained 85.0% (corrected 88.7% in normal samples and corrected 81.2% in defected samples) and 91.7% (corrected 100% in normal samples and corrected 83.9% in defected samples) of the total accuracy for calibration and prediction, respectively. Moreover, defected samples were classified in 3 levels of infection severity. The accuracies of cross validation for groups of low, medium and high infection severity were investigated and obtained 82.2, 82.4 and 90.0%, respectively. In conclusion, the calibration model from transmittance NIRS technique can be applied for rapid and non-destructive sorting of internal mold infection in intact tomato.
Some of the metrics are blocked by your
consent settings
Non-destructive prediction of hardening pericarp disorder in intact mangosteen by near infrared transmittance spectroscopy
(2011-10-01)
Teerachaichayut, Sontisuk
;
Terdwongworakul, Anupun
;
Thanapase, Warunee
;
Kiji, Kazuaki
A non-destructive technique to predict a hardening pericarp disorder in intact mangosteen is proposed by using near infrared (NIR) transmittance spectroscopy in the wavelength range of 660-960 nm. The study found that the spectral features of normal pericarp mangosteen and hardening pericarp mangosteen were different. The averaged spectra and individual spectra of hardening pericarp mangosteen from a calibration set (N = 560) were used to develop classification models, using partial least squares discriminant analysis (PLS-DA). A model based on individual spectra obtained better classification. The overall accuracy of classification for a prediction set (N = 358) was 91%. Out of 179 samples of normal pericarp fruits, 167 were identified correctly, while 159 samples out of 179 samples with hard pericarp were predicted correctly. The results showed that NIR transmittance spectroscopy can be used to predict hard pericarp disorder in intact mangosteen fruit accurately. © 2011 Elsevier Ltd. All rights reserved.
Some of the metrics are blocked by your
consent settings
Microwave sensor response in relation to durian maturity
(2009-12-01)
Suttapa, S.
;
Varith, J.
;
Krairiksh, M.
;
Noochuay, C.
;
Phimpimol, J.
This research studied the microwave response in relation to maturity development of durian fruit (Durio zibethinus Murry cv. Monthong) using a microwave sensor at frequency of 2.45 GHz. The microwave sensor applied two patch antennas to couple the microwave signal from durian. The experiment was conducted on 30 intact durian fruits at 91-123 days after full-bloom. One intact fruit on the tree was attached with the microwave sensor onto durian peel. The microwave sensor response was recorded everyday. Meanwhile, three durian fruits with the same size and age were picked from the tree for every third day to determine their maturity indices e.g., reducing sugar, starch, moisture content, dry matter and dielectric properties of durian pulp. Result showed that the microwave sensor response linearly decreased during the first period day 91-103, then linearly increased during day 104-118. As the wave from microwave sensor penetrated through the peel, it may interact to one of the maturity indices in the pulp, possibly moisture content during maturity of 50- 70% maturity and starch content during maturity of 70-90%. Dielectric constant of durian pulp at 2.45 GHz was found as a confound factor affecting microwave sensor response relating to durian maturity. The transition of microwave response was approximately at day 103, implying that durian maturity was about 70%, corresponding to an onset of reducing sugar. The microwave responses may be beneficial for durian farmers to predict the suitable harvesting period because of its nondestructive characteristic for durian maturity prediction.

KMITL

Browse

Filters

Settings

Sort By

Results per page

Search Results