秋月梨最佳采收期近红外光谱预测：考虑环境光变化和仪器台间差校正

doi:10.3964/j.issn.1000-0593(2025)10-2774-09

摘要
参考文献
相关文章 (15)

全文: PDF (4510 KB)
输出: BibTeX | EndNote (RIS)

摘要：适时采收是提高秋月梨果实采收优质率和贮藏品质的重要科学探索。果园环境光变化和仪器台间差会导致实验室内建立的预测树上果实品质数学模型的性能下降。为了减少果园环境光变化和仪器台间差的影响，采用全局模型、外部参数正交化（EPO）方法对环境光影响和仪器台间差进行校正，并预测秋月梨的最佳采收期。采用2019年到2020年之间采集的599枚秋月梨样品建模，80枚秋月梨作为差异矩阵样品，2023年7月至9月采集的120枚秋月梨作为预测集。经过全局模型和EPO校正后，偏最小二乘回归（PLSR）模型的预测能力均有提高，全局模型最优、EPO次之。经过全局模型校正后，仪器A的模型预测仪器B的数据，决定系数由0.11提高到0.68，可溶性固形物（SSC）的预测均方根误差（RMSEP）由1.2%降低到0.69%。同时进行台间差和环境光变化全局模型校正后，决定系数由0.46提高到0.79，RMSEP由1.23%降低到0.70%。采用全局模型修正后的最佳模型预测最佳采收期，手持式仪器的预测结果与破坏性分析结果一致性较好，实验果园的最佳采收期为2023年8月26日，55%的采样秋月梨SSC含量超12%，满足采收标准。结果表明，全局模型校正能同时有效减小仪器台间差和环境光变化对模型预测性能的影响。本研究验证了采收期预测能提高秋月梨的采收品质，为秋月梨最佳采收期无损预测提供了参考。

关键词：近红外光谱；可溶性固形物含量；手持式光谱检测仪；最佳采收期；光谱差矩阵；外部参数正交化；全局模型

Abstract：Timely harvesting is an important scientific exploration to improve the high-quality rate of autumn moon pear fruit harvest and storage quality. The influence of light changes in the orchard environment and inter-instrument differences can lead to a decline in the performance of the mathematical model established in the laboratory when predicting the quality of fruits on the tree. This study simultaneously considers the influence of light changes in the orchard environment and inter-instrument differences. It utilizes the global model and external parameter orthogonalization (EPO) method to correct for the influence of environmental light and inter-instrument differences, thereby predicting the optimal harvest date of autumn moon pears. The experiment used 599 autumn moon pear samples collected between 2019 and 2020 for modeling, 80 autumn moon pears as difference matrix samples, and 120 autumn moon pears collected from July to September 2023 as the prediction set. After global model and EPO correction, the predictive ability of the partial least squares regression (PLSR) model was improved, with the global model being the best and EPO being the second. After global model correction, the model of instrument A predicted the data of instrument B, and the coefficient of determination increased from 0.11 to 0.68, and the root mean square error of prediction (RMSEP) of soluble solid content (SSC) decreased from 1.2% to 0.69%. After simultaneous correction of inter-instrument differences and ambient light changes using the global model, the coefficient of determination increased from 0.46 to 0.79, and RMSEP decreased from 1.23% to 0.70%. The best model corrected by the global model, was used to predict the optimal harvest date, and the prediction results from the handheld instrument were consistent with the destructive analysis results. The optimal harvest period of the experimental orchard was August 26, 2023, and 55% of the sampled autumn moon pears had an SSC content exceeding 12%, meeting the harvest standard. The results demonstrate that global model correction can effectively mitigate the impact of inter-instrument differences and ambient light changes on the model's predictive performance. At the same time, this study verified that harvest date prediction can improve the harvest quality of autumn moon pears, providing a feasible reference for non-destructive prediction of the optimal harvest date of autumn moon pears.

Key words：Near infrared spectroscopy; Soluble solids content; Handheld spectrometer; Optimum harvest time; Spectral difference matrix; EPO; Global model

收稿日期: 2025-01-06 修订日期: 2025-06-06

中图分类号:

O657.3

基金资助: 江西省自然科学基金项目（20242BAB25361），国家自然科学基金项目（31960497）和江西省职业早期青年项目（20244BCE52165）资助

作者简介: 孙旭东，1978年生，华东交通大学机电与车辆工程学院教授 e-mail: sunxudong_18@163.com

引用本文:

孙旭东，龙涛，王加华，冯绍然，曾体伟，谢冬福，付伟. 秋月梨最佳采收期近红外光谱预测：考虑环境光变化和仪器台间差校正[J]. 光谱学与光谱分析, 2025, 45(10): 2774-2782.
SUN Xu-dong, LONG Tao, WANG Jia-hua, FENG Shao-ran, ZENG Ti-wei, XIE Dong-fu, FU Wei. Near-Infrared Spectroscopy Prediction of the Optimal Harvest Date for Autumn Moon Pear: Considering the Correction of Ambient Light Changes and Inter-Instrument Differences. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(10): 2774-2782.

链接本文:

https://www.gpxygpfx.com/CN/10.3964/j.issn.1000-0593(2025)10-2774-09 或 https://www.gpxygpfx.com/CN/Y2025/V45/I10/2774

[1] General Administration of Quality Supervision, Inspection and Qurantine of the People's Republic of China(中华人民共和国质量监督检验检疫总局). GB/T 10650—2008—Fresh Pear(GB/T 10650—2008—鲜梨), 2008.
[2] Walsh K B, McGlone V A, Han D H. Postharvest Biology and Technology, 2020, 163: 111139.
[3] Flores K, SánchezMT, Pérez-MarínD, Guerrero JE, et al. Journal of Food Engineering, 2009, 91(2): 311.
[4] Walsh K B, Blasco J, Zude-Sasse M, et al. Postharvest Biology and Technology, 2020, 168: 111246.
[5] Minas I S, Blanco-Cipollone F, Sterle D. Food Chemistry, 2021, 335(15): 127626.
[6] Saranwong S, Sornsrivichai J, Kawano S. Journal of Near Infrared Spectroscopy, 2003, 11: 283.
[7] Zude M, Pflanz M, Kaprielian C, et al. Biosystems Engineering, 2008, 99(3): 455.
[8] Kanchanomai C, Ohashi S, Naphrom D, et al. Horticulture, Environment, and Biotechnololgy, 2020, 61(4): 725.
[9] Feng J, McGlone A V, Currie M, et al. HortScience, 2011, 46(1): 57.
[10] Roger J M, Chauchard F, Bellon-Maurel V. Chemometrics and Intelligent Laboratory Systems, 2003, 66(2): 191.
[11] Sun X, Du Y, Nawaz M, et al. Postharvest Biology and Technology, 2024, 217: 113101.
[12] LIU Yan-de, LIAO Jun, LI Bin, et al（刘燕德，廖军，李斌，等）. Spectroscopy and Spectral Analysis（光谱学与光谱分析）, 2022，42(9): 2781.
[13] Sun X, Subedi P, Walker R, et al. Postharvest Biology and Technology, 2020, 163: 111140.
[14] Zude M, Herold B, Roger J M, et al. Journal of Food Engineering, 2006, 77(2): 254.
[15] Peirs A, Lammertyn J, Ooms K, et al. Postharvest Biology and Technology, 2000, 21(2): 189.
[16] Bertone E, Venturello A, Leardi R, et al. Postharvest Biology and Technology, 2012, 69: 15.
[17] Peirs A, Schenk A, Nicola B M. Postharvest Biology and Technology, 2005, 35(1): 1.
[18] Cavaco A M, Antunes M D, Guerra R, et al. Acta Horticulturae, 2018, 1194: 135.
[19] Choi J H, Chen P A, Lee B H, et al. Scientia Horticulturae, 2017, 220: 147.
[20] Williams P. NIR News, 2014, 25(1): 22.
[21] Barnaba F E, Bellincontro A, Mencarelli F. Journal of the Science of Food and Agriculture, 2014, 94(6): 1071.
[22] Afonso A M, Antunes M D, Cruz S, et al. Postharvest Biology and Technology, 2022, 188: 111895.
[23] Goisser S, Fernandes M, Wittmann S, et al. Journal of Applied Botany and Food Quality, 2020, 93: 204.
[24] Nicolai B M, Beullens K, Bobelyn E, et al. Postharvest Biology and Technology, 2007, 46 (2): 99.
[25] Sun X, Guo F, Liu J, et al. Computers and Electronics in Agriculture, 2024, 219: 108797.
[26] Cavaco A M, Pires R, Antunes M D, et al. Postharvest Biology and Technology, 2018, 141: 86.
[27] Sun X, Subedi P, Walsh K B. Postharvest Biology and Technology, 2020, 162: 111117.
[28] Cavaco A M, Passos D, Pires R M, et al. Non Destructive Assessment of Citrus Fruit Quality and Ripenning by Visible-Near Infrared Reflectance Spectroscopy, 2021, doi: 10.5772 intechopen.95970.
[29] Sun X, Wang Z, Aydin H, et al. Postharvest Biology and Technology, 2023, 201: 112374.
[30] Kwon J H, Park H Y, Jun J H, et al. Horticulture Environment & Biotechnology, 2008, 49(2): 85.
[31] Wert T W, Williamson J G, Chaparro J X, et al. Hortscience, 2009, 44(3): 666.
[32] Pratiwi E Z D, Pahlawan M F R, Rahmi D N, et al. Open Agriculture, 2023, 8(1): 946.