Advanced chemometrics toward robust spectral analysis for fruit quality evaluation

被引:20
作者
Zhang, Xiaolei [1 ,2 ]
Yang, Jie [3 ]
机构
[1] Nanjing Agr Univ, Coll Engn, Nanjing 210031, Peoples R China
[2] Sanya Inst Nanjing Agr Univ, Sanya 572025, Peoples R China
[3] Univ Minnesota Twin Cities, Dept Bioprod & Biosyst Engn, St Paul, MN 55108 USA
基金
中国国家自然科学基金;
关键词
Food quality; Postharvest; Visible/near-infrared spectroscopy; Model robustness; Deep learning; Biological variability; SOLUBLE SOLIDS CONTENT; NEAR-INFRARED SPECTROSCOPY; NONDESTRUCTIVE PREDICTION; BIOLOGICAL VARIABILITY; CALIBRATION TRANSFER; NIR SPECTROSCOPY; MEASUREMENT POSITION; FOOD QUALITY; MODELS; APPLES;
D O I
10.1016/j.tifs.2024.104612
中图分类号
TS2 [食品工业];
学科分类号
0832 ;
摘要
Background: The application of visible/near-infrared (Vis/NIR) spectroscopy for fruit quality evaluation has garnered significant attention over the past few decades. Various chemometric techniques have been developed to predict fruit quality from spectral data. However, the broad applicability of existing chemometric models is limited by unpredictable data variability caused by various biological factors, instrumental settings, and measurement conditions. Deep learning has emerged as a leading methodology, offering substantial improvements in the accuracy and robustness of fruit quality assessments. Scope and approach: This review examines the challenges of model robustness in fruit spectral analysis, tracing the advancement from conventional chemometrics to deep learning approaches. Developments in chemometric methods to enhance model reliability are explored, encompassing dataset-level, variable-level, and model parameter-level strategies, while outlining their applicability and limitations. Recent advances in deep learningbased techniques, e.g., transfer learning, multi-task learning, multi-modal data fusion, and knowledge-guided model design, are further highlighted, providing prospective pathways for achieving superior model robustness. Key findings and conclusions: Current chemometric methods have enhanced model accuracy and proven effective in fruit spectral analysis. While the results are improved for certain research objectives, many analyses remain dependent on specific dataset characteristics and manual feature engineering, such as preprocessing, which limits their generalizability. Deep learning techniques with advanced feature extraction capabilities have shown promise in reducing the need for manually engineered features and expanding model robustness. However, further investigation into the applicability and limitations of these models is crucial for their successful integration into chemometric analysis.
引用
收藏
页数:16
相关论文
共 50 条
[31]   Evaluation of Quality Parameters of Acai Oil During Thermal Oxidation Using NIRS and Chemometrics [J].
Rodrigues, B. V. C. ;
Pompeu, D. R. ;
Herman, C. A. N. P. .
FOOD ANALYTICAL METHODS, 2021, 14 (11) :2335-2345
[32]   Comparative analysis of models for robust and accurate evaluation of soluble solids content in 'Pinggu' peaches by hyperspectral imaging [J].
Li, Jiangbo ;
Chen, Liping .
COMPUTERS AND ELECTRONICS IN AGRICULTURE, 2017, 142 :524-535
[33]   Recent Advances in Spectral Analysis Techniques for Non-Destructive Detection of Internal Quality in Watermelon and Muskmelon : A Review [J].
Ma Ben-xue ;
Yu Guo-wei ;
Wang Wen-xia ;
Luo Xiu-zhi ;
Li Yu-jie ;
Li Xiao-zhan ;
Lei Sheng-yuan .
SPECTROSCOPY AND SPECTRAL ANALYSIS, 2020, 40 (07) :2035-2041
[34]   Evaluation of NIRS as non-destructive test to evaluate quality traits of purple passion fruit [J].
Maniwara, Phonkrit ;
Nakano, Kazuhiro ;
Ohashi, Shintaroh ;
Boonyakiat, Danai ;
Seehanam, Pimjai ;
Theanjumpo, Parichat ;
Poonlarp, Pichaya .
SCIENTIA HORTICULTURAE, 2019, 257
[35]   Fruit quality evaluation of plantains, plantain hybrids, and cooking bananas [J].
Ferris, RSB ;
Ortiz, R ;
Vuylsteke, D .
POSTHARVEST BIOLOGY AND TECHNOLOGY, 1999, 15 (01) :73-81
[36]   Assessment of rind quality of 'Nules Clementine' mandarin fruit during postharvest storage: 2. Robust Vis/NIRS PLS models for prediction of physico-chemical attributes [J].
Magwaza, Lembe Samukelo ;
Opara, Umezuruike Linus ;
Cronje, Paul J. R. ;
Landahl, Sandra ;
Nieuwoudt, Helene H. ;
Mouazen, Abdul M. ;
Nicolai, Bart M. ;
Terry, Leon A. .
SCIENTIA HORTICULTURAE, 2014, 165 :421-432
[37]   Advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: A review - Part I: Fundamentals [J].
Wu, Di ;
Sun, Da-Wen .
INNOVATIVE FOOD SCIENCE & EMERGING TECHNOLOGIES, 2013, 19 :1-14
[38]   Advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: A review - Part II: Applications [J].
Wu, Di ;
Sun, Da-Wen .
INNOVATIVE FOOD SCIENCE & EMERGING TECHNOLOGIES, 2013, 19 :15-28
[39]   Advanced Line Sampling for efficient robust reliability analysis [J].
de Angelis, Marco ;
Patelli, Edoardo ;
Beer, Michael .
STRUCTURAL SAFETY, 2015, 52 :170-182
[40]   Rapid quality evaluation of cinnamomi cortex based on near-infrared spectroscopy and chemometrics methods [J].
Shen, Rongjing ;
Sang, Mengjiao ;
Li, Juan ;
Zang, Hengchang ;
Zhang, Zhonghu ;
Nie, Lei .
MICROCHEMICAL JOURNAL, 2025, 214