Online Inspection of Browning in Yali Pears Using Visible-Near Infrared Spectroscopy and Interpretable Spectrogram-Based CNN Modeling
Biosensors · 29.01.2023 · 13 (2) · 203
DOI: 10.3390/bios13020203 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biosensors, online erschienen am 29.01.2023. Verfasst von Yong Hao und 3 weiteren Autor:innen. Beteiligte Einrichtung: School of Mechatronics and Vehicle Engineering, East China Jiaotong University, Nanchang 330013, China. Hauptthemen laut MEDLINE: Pyrus. Die Arbeit wurde bisher 33-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Jiangxi Provincial Natural Science Foundation of China.
Abstract (Originalsprache)
Browning is the most common physiological disease of Yali pears during storage. At the initial stage, browning only occurs in the tissues near the fruit core and cannot be detected from the appearance. The disease, if not identified and removed in time, will seriously undermine the quality and sale of the whole batch of fruit. Therefore, there is an urgent need to explore a method for early diagnosis of the browning in Yali pears. In order to realize the dynamic and online real-time detection of the browning in Yali pears, this paper conducted online discriminant analysis on healthy Yali pears and those with different degrees of browning using visible-near infrared (Vis-NIR) spectroscopy. The experimental results show that the prediction accuracy of the original spectrum combined with a 1D-CNN deep learning model reached 100% for the test sets of browned pears and healthy pears. Features extracted by the 1D-CNN method were converted into images by Gramian angular field (GAF) for PCA visual analysis, showing that deep learning had good performance in extracting features. In conclusion, Vis-NIR spectroscopy combined with the 1D-CNN discriminant model can realize online detection of browning in Yali pears.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Mechatronics and Vehicle Engineering, East China Jiaotong University, Nanchang 330013, China.
- Key Laboratory of Conveyance Equipment of the Ministry of Education, Nanchang 330013, China.
- School of Civil Engineering and Architecture, East China Jiaotong University, Nanchang 330013, China.
Förderung
- National Natural Science Foundation of China
- Jiangxi Provincial Natural Science Foundation of China
Zitieren
APA
Hao, Y., Li, X., Zhang, C., & Lei, Z. (2023). Online Inspection of Browning in Yali Pears Using Visible-Near Infrared Spectroscopy and Interpretable Spectrogram-Based CNN Modeling. Biosensors, 13(2), 203. https://doi.org/10.3390/bios13020203
Vancouver
Hao Y, Li X, Zhang C, Lei Z. Online Inspection of Browning in Yali Pears Using Visible-Near Infrared Spectroscopy and Interpretable Spectrogram-Based CNN Modeling. Biosensors. 2023;13(2):203. doi:10.3390/bios13020203
BibTeX
@article{Hao2023,
author = {Hao, Yong and Li, Xiyan and Zhang, Chengxiang and Lei, Zuxiang},
title = {Online Inspection of Browning in Yali Pears Using Visible-Near Infrared Spectroscopy and Interpretable Spectrogram-Based CNN Modeling},
journal = {Biosensors},
year = {2023},
volume = {13},
number = {2},
pages = {203},
doi = {10.3390/bios13020203}
}
EUVIMED