Developing deep learning based regression approaches for determination of chemical compositions in dry black goji berries (Lycium ruthenicum Murr.) using near-infrared hyperspectral imaging

•Black goji berries total anthocyanins, flavonoids and phenolics were predicted by HSI.•SPA, CARS and PCA, WT were compared to obtain conventional features for prediction.•CNN models for regression were proposed and compared.•CNN and DAE were used and compared as feature extraction methods for predi...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Food chemistry Jg. 319; S. 126536
Hauptverfasser: Zhang, Chu, Wu, Wenyan, Zhou, Lei, Cheng, Huan, Ye, Xingqian, He, Yong
Format: Journal Article
Sprache:Englisch
Veröffentlicht: England Elsevier Ltd 30.07.2020
Schlagworte:
anthocyanins
Black goji berry
chemical composition
Convolutional neural network
Deep autoencoder
hyperspectral imagery
least squares
Lycium ruthenicum
Near-infrared hyperspectral imaging
neural networks
nutritive value
phenolic compounds
principal component analysis
Regression issue
selection methods
support vector machines
Total anthocyanins
Total flavonoids
Total phenolics
wavelengths
wavelet
Regression issue
Total flavonoids
Black goji berry
Total anthocyanins
Convolutional neural network
Near-infrared hyperspectral imaging
Total phenolics
Deep autoencoder
ISSN:0308-8146, 1873-7072, 1873-7072
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Black goji berries total anthocyanins, flavonoids and phenolics were predicted by HSI.•SPA, CARS and PCA, WT were compared to obtain conventional features for prediction.•CNN models for regression were proposed and compared.•CNN and DAE were used and compared as feature extraction methods for prediction.•Proposed deep learning approaches could be used to determine chemical compositions. Black goji berry (Lycium ruthenicum Murr.) has great commercial and nutritional values. Near-infrared hyperspectral imaging (NIR-HSI) was used to determine total phenolics, total flavonoids and total anthocyanins in dry black goji berries. Convolutional neural networks (CNN) were designed and developed to predict the chemical compositions. These CNN models and deep autoencoder were used as supervised and unsupervised feature extraction methods, respectively. Partial least squares (PLS) and least-squares support vector machine (LS-SVM) as modelling methods, successive projections algorithm and competitive adaptive reweighted sampling (CARS) as wavelength selection methods, and principal component analysis (PCA) and wavelet transform (WT) as feature extraction methods were studied as conventional approaches for comparison. Deep learning approaches as modelling methods and feature extraction methods obtained good and equivalent performances to the conventional methods. The results illustrated that deep learning had great potential as modelling and feature extraction methods for chemical compositions determination in NIR-HSI.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0308-8146
1873-7072
1873-7072
DOI:10.1016/j.foodchem.2020.126536