Ripeness Classification of Bananito Fruit ( Musa acuminata, AA): a Comparison Study of Visible Spectroscopy and Hyperspectral Imaging

The Rapid Non-Destructive Differentiation of Different Varieties of Rice by Fluorescence Hyperspectral Technology Combined with Machine Learning

A rice classification method for the fast and non-destructive differentiation of different varieties is significant in research at present. In this study, fluorescence hyperspectral technology combined with machine learning techniques was used to distinguish five rice varieties by analyzing the fluorescence hyperspectral features of Thai jasmine rice and four rice varieties with a similar appearance to Thai jasmine rice in the wavelength range of 475-1000 nm. The fluorescence hyperspectral data were preprocessed by a first-order derivative (FD) to reduce the background and baseline drift effects of the rice samples. Then, a principal component analysis (PCA) and t-distributed stochastic neighborhood embedding (t-SNE) were used for feature reduction and 3D visualization display. A partial least squares discriminant analysis (PLS-DA), BP neural network (BP), and random forest (RF) were used to build the rice classification models. The RF classification model parameters were optimized using the gray wolf algorithm (GWO). The results show that FD-t-SNE-GWO-RF is the best model for rice classification, with accuracy values of 99.8% and 95.3% for the training and test sets, respectively. The fluorescence hyperspectral technique combined with machine learning is feasible for classifying rice varieties.

A data fusion approach for nondestructive tracking of the ripening process and quality attributes of green Hayward kiwifruit using artificial olfaction and proximal hyperspectral imaging techniques

A data fusion strategy based on hyperspectral imaging (HSI) and electronic nose (e-nose) systems was developed in this study to inspect the postharvest ripening process of Hayward kiwifruit. The extracted features from the e-nose and HSI techniques, in single or combined mode, were used to develop machine learning algorithms. Performance evaluations proved that the fusion of olfactory and reflectance data improves the performance of discriminative and predictive algorithms. Accordingly, with high classification accuracies of 100% and 94.44% in the calibration and test stages, the data fusion-based support vector machine (SVM) outperformed the partial least square discriminant analysis (PLSDA) for discriminating the kiwifruit samples into eight classes based on storage time. Moreover, the data fusion-based support vector regression (SVR) was a better predictor than partial least squares regression (PLSR) for kiwifruit firmness, soluble solids content (SSC), and titratable acidity (TA) measures. The prediction R 2 and RMSE criteria of the SVR algorithm on the test data were 0.962 and 0.408 for firmness, 0.964 and 0.337 for SSC, and 0.955 and 0.039 for TA, respectively. It was concluded that the hybrid of e-nose and HSI systems coupled with the SVM algorithm delivers an effective tool for accurate and nondestructive monitoring of kiwifruit quality during storage.

Non-destructive pre-symptomatic detection of gray mold infection in kiwifruit using hyperspectral data and chemometrics

Application of hyperspectral imaging (HSI) and data analysis algorithms was investigated for early and non-destructive detection of Botrytis cinerea infection. Hyperspectral images were collected from laboratory-based contaminated and non-contaminated fruits at different day intervals. The spectral wavelengths of 450 nm to 900 nm were pretreated by applying moving window smoothing (MWS), standard normal variates (SNV), multiplicative scatter correction (MSC), Savitzky-Golay 1^st derivative, and Savitzky-Golay 2^nd derivative algorithms. In addition, three different wavelength selection algorithms, namely; competitive adaptive reweighted sampling (CARS), uninformative variable elimination (UVE), and successive projection algorithm (SPA), were executed on the spectra to invoke the most informative wavelengths. The linear discriminant analysis (LDA), developed with SNV-filtered spectral data, was the most accurate classifier to differentiate the contaminated and non-contaminated kiwifruits with accuracies of 96.67% and 96.00% in the cross-validation and evaluation stages, respectively. The system was able to detect infected samples before the appearance of disease symptoms. Results also showed that the gray-mold infection significantly influenced the kiwifruits' firmness, soluble solid content (SSC), and titratable acidity (TA) attributes. Moreover, the Savitzky-Golay 1^st derivative-CARS-PLSR model obtained the highest prediction rate for kiwifruit firmness, SSC, and TA with the determination coefficient (R²) values of 0.9879, 0.9644, 0.9797, respectively, in calibration stage. The corresponding cross-validation R² values were equal to 0.9722, 0.9317, 0.9500 for firmness, SSC, and TA, respectively. HSI and chemometric analysis demonstrated a high potential for rapid and non-destructive assessments of fungal-infected kiwifruits during storage.

Characterising Colour Feature Descriptors for Ficus carica L. Ripeness Classification Based on Artificial Neural Network (ANN)

Excessive feature dimensions impact the effectiveness of machine learning, computationally expensive and the analysis of feature correlations in the engineering area. This paper uses the colour descriptor to get the most optimal feature to improve time consumption and efficiency. This study investigated Ficus carica L. (figs) with three classification stages. The ripening classification of fig was examined using colour features descriptor with two different colour models, RGB and HSV. In addition, the machine learning classification model based on Artificial Neural Network (ANN) that utilised the Feed-Forward Neural Network (FFNN) model to classify the ripeness of fig is considered in this characterisation. Five different numbers of binning were characterised for RGB and HSV. Both colour feature descriptors were compared in terms of accuracy, sensitivity, precision, and time consumption to identify the dimension of the optimal feature. Based on the result, reducing the size of images will improve the time consumption with comparable accuracy. Moreover, the reduction of features dimension cannot be too small or too big due to inequitable enough to differentiate the ripeness stages and lead to a false error state. The optimal features dimension in binning for RGB was 8 (R/G/B) bins with 96.7% accuracy. Meanwhile, 96.7% accuracy for HSV at 15, 5, and 5 (H, S, V) bins as optimal colour features.

CRISPR/Cas14a-Based Isothermal Amplification for Profiling Plant MicroRNAs

MicroRNAs (miRNAs) play key roles in biological processes in plants, such as stress resistance, yet can hardly be quantified by an enzyme-involved terminal polymerization process due to their 2'-O-methyl modifications at the 3'-terminal. Herein, we proposed a CRISPR/Cas14a-based rolling circle amplification (termed Cas14R) assay, allowing reverse transcription-free and demethylation-free detection of plant miRNAs with single-nucleotide resolution. The employment of target-templated rolling circle amplification circumvents the extension of the unaccessible 2'-O-methyl group at the 3'-terminal. Particularly, the activated Cas14a confers the trans-cleavage activity for identifying target single-stranded DNA sequences without the necessity of the protospacer adjacent motif, generalizing the detection of miRNA sequences and the integration of different isothermal amplification techniques. Ultimately, the Cas14R assay has been applied to profile miR156a to evaluate the ripeness process of banana, indicating its feasibility in analyzing the roles of miRNAs in biological processes of plants.

Evaluating drying feature differences between ginger slices and splits during microwave-vacuum drying by hyperspectral imaging technique

This study aimed to investigate the difference between ginger slices (vertically cut) and splits (horizontally cut) during microwave-vacuum drying (MVD) procedures. MVD ginger slices showed a higher shrinkage rate and a higher hardness value, with a more porous structure of the surface layer. MVD ginger splits had higher rehydration rates at the first 15 min of the rehydration. Nine optimal wavelengths were selected by regression coefficients (RC) from the partial least squares regression (PLSR) model based on the raw data. A simplified PLSR model based on optimal wavelengths showed a good performance with a coefficient of determination in prediction (Rp2) of 0.973 and a root mean square error in prediction (RMSEP) of 4.63%. Texture features of grey level co-occurrence matrix (GLCM) of moisture prediction maps demonstrated a more uniform moisture distribution in MVD ginger slices than that in splits in the original geometry.

Application and interpretation of deep learning methods for the geographical origin identification of Radix Glycyrrhizae using hyperspectral imaging

Radix Glycyrrhizae is used as a functional food and traditional medicine. The geographical origin of Radix Glycyrrhizae is a determinant factor influencing the chemical and physical properties as well as its medicinal and health effects. The visible/near-infrared (Vis/NIR) (376–1044 nm) and near-infrared (NIR) hyperspectral imaging (915–1699 nm) were used to identify the geographical origin of Radix Glycyrrhizae. Convolutional neural network (CNN) and recurrent neural network (RNN) models in deep learning methods were built using extracted spectra, with logistic regression (LR) and support vector machine (SVM) models as comparisons. For both spectral ranges, the deep learning methods, LR and SVM all exhibited good results. The classification accuracy was over 90% for the calibration, validation, and prediction sets by the LR, CNN, and RNN models. Slight differences in classification performances existed between the two spectral ranges. Further, interpretation of the CNN model was conducted to identify the important wavelengths, and the wavelengths with high contribution rates that affected the discriminant analysis were consistent with the spectral differences. Thus, the overall results illustrate that hyperspectral imaging with deep learning methods can be used to identify the geographical origin of Radix Glycyrrhizae, which provides a new basis for related research.

Hyperspectral imaging provides an effective way to identify the geographical origin of Radix Glycyrrhizae to assess its quality.