Опубликована: Янв. 1, 2024
Язык: Английский
Опубликована: Янв. 1, 2024
Язык: Английский
Agricultural Research, Год журнала: 2025, Номер unknown
Опубликована: Фев. 10, 2025
Язык: Английский
Процитировано
2Food Bioscience, Год журнала: 2024, Номер unknown, С. 105558 - 105558
Опубликована: Ноя. 1, 2024
Язык: Английский
Процитировано
3Journal of the Taiwan Institute of Chemical Engineers, Год журнала: 2025, Номер 171, С. 106045 - 106045
Опубликована: Фев. 27, 2025
Язык: Английский
Процитировано
0Trends in Food Science & Technology, Год журнала: 2025, Номер unknown, С. 104977 - 104977
Опубликована: Март 1, 2025
Язык: Английский
Процитировано
0Food Research International, Год журнала: 2025, Номер unknown, С. 116285 - 116285
Опубликована: Март 1, 2025
Язык: Английский
Процитировано
0Food Science & Nutrition, Год журнала: 2025, Номер 13(4)
Опубликована: Март 27, 2025
ABSTRACT In this research, the intelligent quality grading of green cardamom was carried out using electronic nose (e‐nose) and computer vision (CV) methods along with machine learning (ML) approaches. Cardamom samples were analyzed in three grades including Grade 1 (healthy green), 2 yellow color), 3 (immature shriveled) for capsules (Black), (Brown), (Yellow red) seeds. Three ML algorithms Decision Tree (DT), Bayesian Network (BN), Support Vector Machine (SVM) used to classify grades. Results showed that correlation‐based feature selection (CFS) algorithm decreased number input features increased classification performance. For classifying capsule based on visual features, CFS‐BN model best classifier, root mean squared error (RMSE) accuracy 0.1408 96.67%, respectively. The RMSE seeds image 0.1220 e‐nose data, CFS‐DT classifier 0.2093 93.33%, an 0.1126 96.67%. fusion CV data performance compared separate use datasets. combination 100% during both calibration evaluation stages. It can be concluded effectively develop intelligent, accurate, reliable, fast, non‐destructive system
Язык: Английский
Процитировано
0Journal of Food Measurement & Characterization, Год журнала: 2025, Номер unknown
Опубликована: Апрель 2, 2025
Язык: Английский
Процитировано
0Flavour and Fragrance Journal, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 25, 2024
ABSTRACT Coffee, a popular beverage worldwide, requires thorough quality assessment to ensure its authenticity and meet consumer demands. Traditional methods in the industry are often subjective, expensive, time‐consuming. This study used compact, portable electronic nose (e‐nose) with machine learning models classify distinguish between civet non‐civet roasted beans. The polynomial feature extraction method was extract important parameters from sensor response improve system performance. Classification like linear discriminant analysis (LDA), logistic regression (LR), quadratic (QDA), support vector machines (SVM) were applied samples. Among these, LDA model features yielded highest validation test accuracies, values of 0.89 ± 0.04 0.93, respectively. higher than statistical methods, which obtained accuracies 0.80 0.07 0.87, acquired e‐nose results correlated compound concentrations coffee beans measured by gas chromatography–mass spectrometry (GC–MS). These findings demonstrate system's promising potential effectively based on their aroma profiles using methods.
Язык: Английский
Процитировано
1Опубликована: Янв. 1, 2024
Язык: Английский
Процитировано
0