EUVIMEDOpdag EUVIMED
Crossref

Monitoring green tea fixation quality by intelligent sensors: comparison of image and spectral information

Yuyu Chen iD, Huiting Wu, Ying Liu, Yujie Wang, Chengye Lu, Tiehan Li iD, Yuming Wei, Jingming Ning iD

Journal of the Science of Food and Agriculture · 2023-04 · 103 (6) · 3093-3101

DOI: 10.1002/jsfa.12350 ↗

At a glance

Journal article in Journal of the Science of Food and Agriculture, published online on 2022-12-07. Written by Yuyu Chen and 7 co-authors. Institution: State Key Laboratory of Tea Plant Biology and Utilization, Anhui Agricultural University, Hefei, China. Major MEDLINE topics: Tea, Spectroscopy, Near-Infrared. Cited 15 times so far (Crossref); 37 references. Funded by: National Key Research and Development Program of China.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

BACKGROUND Intelligent monitoring of fixation quality is a prerequisite for automated green tea processing. To meet the requirements of intelligent monitoring of fixation quality in large‐scale production, fast and non‐destructive detection means are urgently needed. Here, smartphone‐coupled micro near‐infrared spectroscopy and a self‐built computer vision system were used to perform rapid detection of the fixation quality in green tea processing lines. RESULTS Spectral and image information from green tea samples with different fixation degrees were collected at‐line by two intelligent monitoring sensors. Competitive adaptive reweighted sampling and correlation analysis were employed to select feature variables from spectral and color information as the target data for modeling, respectively. The developed least squares support vector machine (LS‐SVM) model by spectral information and the LS‐SVM model by image information achieved the best discriminations of sample fixation degree, with both prediction set accuracies of 100%. Compared to the spectral information, the image information‐based support vector regression model performed better in moisture prediction, with a correlation coefficient of prediction of 0.9884 and residual predictive deviation of 6.46. CONCLUSION The present study provided a rapid and low‐cost means of monitoring fixation quality, and also provided theoretical support and technical guidance for the automation of the green tea fixation process. © 2022 Society of Chemical Industry.

Subjects

MeSH terms (MEDLINE)

TeaSpectroscopy, Near-InfraredLeast-Squares AnalysisSupport Vector Machine

Author keywords

computer vision systemfixation degreegreen teamicro-NIR spectroscopymoisture contentsensory evaluation

Substances

Tea

Authors and institutions

  • State Key Laboratory of Tea Plant Biology and Utilization, Anhui Agricultural University, Hefei, China.

ORCID profiles: Yuyu Chen, Tiehan Li, Jingming Ning

Funding

  • National Key Research and Development Program of China

Cite

APA

Chen, Y., Wu, H., Liu, Y., Wang, Y., Lu, C., Li, T., Wei, Y., & Ning, J. (2023). Monitoring green tea fixation quality by intelligent sensors: comparison of image and spectral information. Journal of the Science of Food and Agriculture, 103(6), 3093-3101. https://doi.org/10.1002/jsfa.12350

Vancouver

Chen Y, Wu H, Liu Y, Wang Y, Lu C, Li T, et al. Monitoring green tea fixation quality by intelligent sensors: comparison of image and spectral information. Journal of the Science of Food and Agriculture. 2023;103(6):3093-3101. doi:10.1002/jsfa.12350

BibTeX
@article{Chen2023,
  author = {Chen, Yuyu and Wu, Huiting and Liu, Ying and Wang, Yujie and Lu, Chengye and Li, Tiehan and Wei, Yuming and Ning, Jingming},
  title = {Monitoring green tea fixation quality by intelligent sensors: comparison of image and spectral information},
  journal = {Journal of the Science of Food and Agriculture},
  year = {2023},
  volume = {103},
  number = {6},
  pages = {3093--3101},
  doi = {10.1002/jsfa.12350}
}
Start for free →