Quality analysis of Rubus chingii Hu. based on near-infrared technology

doi:10.16178/j.issn.0528-9017.20250226

Abstract

Abstract:

To establish a quantitative prediction model based on near-infrared spectroscopy data, and to realize the rapid prediction of ellagic acid and kaempferol-3-O-rutinoside content in Rubus chingii Hu., the samples from different sources were collected, the near-infrared spectroscopy was collected, and the contents of ellagic acid and kaempferol-3-O-rutinoside were detected by HPLC. Spectral data were preprocessed by MATLAB R2020b software. The competitive adaptive reweighted sampled CARS were used to screen characteristic wavelengths, and partial least squares (PLS) and random forest (RF) models were established to screen the best pretreatment methods, and then the best prediction models were selected. Results showed that the optimal prediction results of ellagic acid were obtained by SNV+FD+CARS+PLS model, and the correlation coefficient ($R_{\mathrm{p}}^{2}$) of the quantitative model test set was 0.903 8. The optimal prediction results of kaempferol-3-O-rutinoside were obtained by SG+FD+CARS+PLS model, and the $R_{\mathrm{p}}^{2}$ of the quantitative model test set was 0.758 6. The NIR data treated by the combination of normalization and first derivative can distinguish Rubus chingii Hu. qualified products in OPLS-DA model, the cumulative variance value R²Y was 0.728, and the prediction rate Q² was 0.681. The results of this study indicated that the content of ellagic acid and kaempferol-3-O-rutinoside in Rubus chingii Hu. can be quickly predicted by NIR technique, and the quality of raspberry samples can be preliminarily judged.

Key words: Rubus chingii Hu., ellagic acid, kaempferol-3-O-rutinoside, near infrared spectrum

CLC Number:

S567.9

WANG Chuanbao, HU Chunli, ZHAN Jianyong, LI Haishen, JIANG Ling, ZHENG Pinghan, WANG Zhian, SUN Jian. Quality analysis of Rubus chingii Hu. based on near-infrared technology[J]. Journal of Zhejiang Agricultural Sciences, 2025, 66(9): 2126-2136.

Figures/Tables 10

Fig.1 Content distribution of ellagic acid and kaempferol-3-O-rutinoside in Rubus chingii Hu.

Fig.2 Original near-infrared spectrum of Rubus chingii Hu.samples

Table 1 Allocation of Rubus chingii Hu. sample groups

类别	样本	样本数量/个	最大值/%	最小值/%	平均值/%	标准偏差
鞣花酸	整体	80	0.516 0	0.047 0	0.177 9	0.101 8
	训练集	56	0.516 0	0.052 3	0.177 7	0.104 0
	测试集	24	0.426 3	0.047 0	0.178 2	0.098 7
山柰酚-3-O-芸香糖苷	整体	80	0.203 1	0.022 3	0.076 9	0.037 1
	训练集	56	0.203 1	0.035 8	0.089 6	0.037 1
	测试集	24	0.064 1	0.022 3	0.047 2	0.011 3

Fig.3 OPLS-DA distribution maps of different near-infrared preprocessing methods

Table 2 Prediction results of ellagic acid under different data preprocessing methods

模型	预处理方法	训练集			测试集
模型	预处理方法	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
RF	Raw	0.498 9	0.070 6	1.425 4	0.116 6	0.096 5	1.086 8
	NDH	0.645 0	0.063 6	1.693 6	0.413 4	0.064 5	1.333 7
	NDH+FD	0.805 5	0.049 2	2.288 0	0.577 3	0.042 5	1.571 2
	MSC	0.777 6	0.047 6	2.139 8	0.352 1	0.074 5	1.269 1
	MSC+FD	0.635 9	0.057 2	1.672 2	0.599 5	0.069 5	1.614 1
	SG	0.500 8	0.073 7	1.428 1	0.124 3	0.086 3	1.091 6
	SG+FD	0.754 9	0.050 9	2.038 0	0.283 4	0.081 9	1.206 7
	SNV	0.706 2	0.052 8	1.861 7	0.317 1	0.086 8	1.236 1
	SNV+FD	0.781 6	0.047 2	2.159 2	0.413 4	0.076 7	1.333 8
	FD	0.798 2	0.043 6	2.246 0	0.418 8	0.084 1	1.339 9
	SD	0.814 5	0.043 7	2.342 9	0.201 3	0.089 2	1.143 0
PLS	Raw	0.587 9	0.066 6	1.571 9	0.565 3	0.059 8	1.549 4
	NDH	0.627 3	0.063 0	1.652 9	0.400 4	0.074 4	1.319 2
	NDH+FD	0.889 9	0.030 6	3.041 1	0.662 3	0.069 3	1.757 8
	MSC	0.680 0	0.045 8	1.783 8	0.536 1	0.090 8	1.499 8
	MSC+FD	0.929 4	0.024 7	3.797 9	0.585 2	0.075 7	1.586 0
	SG	0.555 0	0.065 8	1.512 7	0.403 1	0.082 4	1.314 7
	SG+FD	0.744 6	0.052 5	1.996 8	0.259 3	0.081 2	1.186 9
	SNV	0.715 4	0.052 0	1.891 6	0.396 9	0.083 5	1.315 3
	SNV+FD	0.899 7	0.030 9	3.186 6	0.735 9	0.055 0	1.987 8
	FD	0.905 0	0.034 2	3.274 3	0.302 6	0.051 4	1.223 2
	SD	0.902 9	0.031 7	3.237 9	0.520 6	0.067 6	1.475 3

Table 2 Prediction results of ellagic acid under different data preprocessing methods

模型	预处理方法	训练集			测试集
模型	预处理方法	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
RF	Raw	0.498 9	0.070 6	1.425 4	0.116 6	0.096 5	1.086 8
	NDH	0.645 0	0.063 6	1.693 6	0.413 4	0.064 5	1.333 7
	NDH+FD	0.805 5	0.049 2	2.288 0	0.577 3	0.042 5	1.571 2
	MSC	0.777 6	0.047 6	2.139 8	0.352 1	0.074 5	1.269 1
	MSC+FD	0.635 9	0.057 2	1.672 2	0.599 5	0.069 5	1.614 1
	SG	0.500 8	0.073 7	1.428 1	0.124 3	0.086 3	1.091 6
	SG+FD	0.754 9	0.050 9	2.038 0	0.283 4	0.081 9	1.206 7
	SNV	0.706 2	0.052 8	1.861 7	0.317 1	0.086 8	1.236 1
	SNV+FD	0.781 6	0.047 2	2.159 2	0.413 4	0.076 7	1.333 8
	FD	0.798 2	0.043 6	2.246 0	0.418 8	0.084 1	1.339 9
	SD	0.814 5	0.043 7	2.342 9	0.201 3	0.089 2	1.143 0
PLS	Raw	0.587 9	0.066 6	1.571 9	0.565 3	0.059 8	1.549 4
	NDH	0.627 3	0.063 0	1.652 9	0.400 4	0.074 4	1.319 2
	NDH+FD	0.889 9	0.030 6	3.041 1	0.662 3	0.069 3	1.757 8
	MSC	0.680 0	0.045 8	1.783 8	0.536 1	0.090 8	1.499 8
	MSC+FD	0.929 4	0.024 7	3.797 9	0.585 2	0.075 7	1.586 0
	SG	0.555 0	0.065 8	1.512 7	0.403 1	0.082 4	1.314 7
	SG+FD	0.744 6	0.052 5	1.996 8	0.259 3	0.081 2	1.186 9
	SNV	0.715 4	0.052 0	1.891 6	0.396 9	0.083 5	1.315 3
	SNV+FD	0.899 7	0.030 9	3.186 6	0.735 9	0.055 0	1.987 8
	FD	0.905 0	0.034 2	3.274 3	0.302 6	0.051 4	1.223 2
	SD	0.902 9	0.031 7	3.237 9	0.520 6	0.067 6	1.475 3

Table 3 Prediction results of kaempferol-3-O-rutinoside under different data preprocessing methods

模型	预处理方法	训练集			测试集
模型	预处理方法	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
RF	Raw	0.559 3	0.023 1	1.520 0	0.142 4	0.037 0	1.103 1
	NDH	0.711 4	0.019 4	1.878 3	0.252 4	0.033 5	1.181 4
	NDH+FD	0.737 2	0.019 7	1.968 2	0.527 7	0.022 6	1.486 4
	MSC	0.705 8	0.020 3	1.860 5	0.258 8	0.030 5	1.186 5
	MSC+FD	0.747 2	0.018 9	2.007 0	0.267 7	0.030 4	1.193 7
	SG	0.676 1	0.017 6	1.773 1	0.195 1	0.041 2	1.138 6
	SG+FD	0.761 1	0.017 3	2.064 5	0.466 8	0.029 4	1.398 9
	SNV	0.676 6	0.021 3	1.774 3	0.203 3	0.032 0	1.144 4
	SNV+FD	0.720 5	0.018 1	1.908 5	0.443 8	0.031 8	1.369 8
	FD	0.752 8	0.019 7	2.029 4	0.376 3	0.022 2	1.293 5
	SD	0.785 1	0.015 8	2.176 7	0.191 1	0.038 1	1.135 8
PLS	Raw	0.414 3	0.022 9	1.318 5	0.402 2	0.037 8	1.321 2
	NDH	0.512 5	0.024 9	1.445 2	0.204 4	0.034 9	1.145 2
	NDH+FD	0.919 7	0.009 1	3.560 1	0.435 5	0.033 7	1.359 6
	MSC	0.534 5	0.027 9	1.497 0	0.185 1	0.022 8	1.131 6
	MSC+FD	0.924 8	0.010 0	3.678 4	0.226 3	0.033 2	1.161 3
	SG	0.413 9	0.029 3	1.318 0	0.216 8	0.027 4	1.154 3
	SG+FD	0.899 8	0.010 4	3.187 0	0.585 7	0.028 0	1.587 0
	SNV	0.527 6	0.027 9	1.468 1	0.299 3	0.020 8	1.220 3
	SNV+FD	0.888 4	0.011 4	3.021 1	0.546 9	0.028 7	1.517 5
	FD	0.867 3	0.012 8	2.769 8	0.454 2	0.030 3	1.382 6
	SD	0.871 7	0.013 2	2.816 6	0.475 2	0.026 7	1.410 1

Table 3 Prediction results of kaempferol-3-O-rutinoside under different data preprocessing methods

模型	预处理方法	训练集			测试集
模型	预处理方法	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
RF	Raw	0.559 3	0.023 1	1.520 0	0.142 4	0.037 0	1.103 1
	NDH	0.711 4	0.019 4	1.878 3	0.252 4	0.033 5	1.181 4
	NDH+FD	0.737 2	0.019 7	1.968 2	0.527 7	0.022 6	1.486 4
	MSC	0.705 8	0.020 3	1.860 5	0.258 8	0.030 5	1.186 5
	MSC+FD	0.747 2	0.018 9	2.007 0	0.267 7	0.030 4	1.193 7
	SG	0.676 1	0.017 6	1.773 1	0.195 1	0.041 2	1.138 6
	SG+FD	0.761 1	0.017 3	2.064 5	0.466 8	0.029 4	1.398 9
	SNV	0.676 6	0.021 3	1.774 3	0.203 3	0.032 0	1.144 4
	SNV+FD	0.720 5	0.018 1	1.908 5	0.443 8	0.031 8	1.369 8
	FD	0.752 8	0.019 7	2.029 4	0.376 3	0.022 2	1.293 5
	SD	0.785 1	0.015 8	2.176 7	0.191 1	0.038 1	1.135 8
PLS	Raw	0.414 3	0.022 9	1.318 5	0.402 2	0.037 8	1.321 2
	NDH	0.512 5	0.024 9	1.445 2	0.204 4	0.034 9	1.145 2
	NDH+FD	0.919 7	0.009 1	3.560 1	0.435 5	0.033 7	1.359 6
	MSC	0.534 5	0.027 9	1.497 0	0.185 1	0.022 8	1.131 6
	MSC+FD	0.924 8	0.010 0	3.678 4	0.226 3	0.033 2	1.161 3
	SG	0.413 9	0.029 3	1.318 0	0.216 8	0.027 4	1.154 3
	SG+FD	0.899 8	0.010 4	3.187 0	0.585 7	0.028 0	1.587 0
	SNV	0.527 6	0.027 9	1.468 1	0.299 3	0.020 8	1.220 3
	SNV+FD	0.888 4	0.011 4	3.021 1	0.546 9	0.028 7	1.517 5
	FD	0.867 3	0.012 8	2.769 8	0.454 2	0.030 3	1.382 6
	SD	0.871 7	0.013 2	2.816 6	0.475 2	0.026 7	1.410 1

Fig.4 CARS wavelength spectral selection diagrams of ellagic acid and kaempferol-3-O-rutinoside from Rubus chingii Hu.

Table 4 Prediction results of ellagic acid under different data preprocessing methods after the CARS algorithm processing

模型	预处理方法	变量/ 个	训练集			测试集
模型	预处理方法	变量/ 个	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
CARS-RF	NDH+FD+CARS	41	0.804 1	0.044 0	2.279 5	0.363 4	0.084 0	1.280 3
	MSC+FD+CARS	36	0.769 1	0.042 6	2.099 7	0.467 7	0.090 1	1.400 1
	SG+FD+CARS	8	0.627 0	0.067 1	1.652 2	0.375 0	0.060 4	1.292 1
	SNV+FD+CARS	28	0.768 3	0.049 6	2.096 2	0.516 5	0.065 4	1.469 1
CARS-PLS	NDH+FD+CARS	41	0.918 3	0.028 9	3.530 2	0.815 5	0.043 2	2.378 4
	MSC+FD+CARS	36	0.914 8	0.028 7	3.456 4	0.810 0	0.046 7	2.343 8
	SG+FD+CARS	8	0.654 7	0.061 9	1.717 1	0.559 1	0.059 5	1.538 4
	SNV+FD+CARS	28	0.930 7	0.025 9	3.833 1	0.903 8	0.032 6	3.292 6

Table 4 Prediction results of ellagic acid under different data preprocessing methods after the CARS algorithm processing

模型	预处理方法	变量/ 个	训练集			测试集
模型	预处理方法	变量/ 个	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
CARS-RF	NDH+FD+CARS	41	0.804 1	0.044 0	2.279 5	0.363 4	0.084 0	1.280 3
	MSC+FD+CARS	36	0.769 1	0.042 6	2.099 7	0.467 7	0.090 1	1.400 1
	SG+FD+CARS	8	0.627 0	0.067 1	1.652 2	0.375 0	0.060 4	1.292 1
	SNV+FD+CARS	28	0.768 3	0.049 6	2.096 2	0.516 5	0.065 4	1.469 1
CARS-PLS	NDH+FD+CARS	41	0.918 3	0.028 9	3.530 2	0.815 5	0.043 2	2.378 4
	MSC+FD+CARS	36	0.914 8	0.028 7	3.456 4	0.810 0	0.046 7	2.343 8
	SG+FD+CARS	8	0.654 7	0.061 9	1.717 1	0.559 1	0.059 5	1.538 4
	SNV+FD+CARS	28	0.930 7	0.025 9	3.833 1	0.903 8	0.032 6	3.292 6

Table 5 Prediction results of kaempferol-3-O-rutinoside under different data preprocessing methods after the CARS algorithm processing

模型	预处理方法	变量/ 个	训练集			测试集
模型	预处理方法	变量/ 个	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
CARS-RF	NDH+FD+CARS	15	0.722 1	0.018 6	1.914 1	0.397 4	0.031 3	1.315 9
	MSC+FD+CARS	18	0.702 7	0.021 5	1.850 7	0.625 3	0.018 3	1.668 8
	SG+FD+CARS	21	0.759 1	0.019 6	2.056 0	0.492 0	0.020 3	1.433 2
	SNV+FD+CARS	34	0.732 4	0.019 3	1.950 7	0.477 9	0.026 1	1.413 7
CARS-PLS	NDH+FD+CARS	15	0.821 0	0.015 5	2.384 8	0.749 9	0.024 3	2.057 3
	MSC+FD+CARS	18	0.817 5	0.014 8	2.361 7	0.598 6	0.026 4	1.612 2
	SG+FD+CARS	21	0.849 7	0.012 2	2.602 5	0.758 6	0.023 2	2.078 9
	SNV+FD+CARS	34	0.838 2	0.014 4	2.508 8	0.752 4	0.019 3	2.052 8

Table 5 Prediction results of kaempferol-3-O-rutinoside under different data preprocessing methods after the CARS algorithm processing

模型	预处理方法	变量/ 个	训练集			测试集
模型	预处理方法	变量/ 个	$R c 2$	RMSEC	RPD	$R p 2$	RMSEP	RPD
CARS-RF	NDH+FD+CARS	15	0.722 1	0.018 6	1.914 1	0.397 4	0.031 3	1.315 9
	MSC+FD+CARS	18	0.702 7	0.021 5	1.850 7	0.625 3	0.018 3	1.668 8
	SG+FD+CARS	21	0.759 1	0.019 6	2.056 0	0.492 0	0.020 3	1.433 2
	SNV+FD+CARS	34	0.732 4	0.019 3	1.950 7	0.477 9	0.026 1	1.413 7
CARS-PLS	NDH+FD+CARS	15	0.821 0	0.015 5	2.384 8	0.749 9	0.024 3	2.057 3
	MSC+FD+CARS	18	0.817 5	0.014 8	2.361 7	0.598 6	0.026 4	1.612 2
	SG+FD+CARS	21	0.849 7	0.012 2	2.602 5	0.758 6	0.023 2	2.078 9
	SNV+FD+CARS	34	0.838 2	0.014 4	2.508 8	0.752 4	0.019 3	2.052 8

Fig.5 Scatter plots of the measured and predicted values of ellagic acid and kaempferol-3-O-rutinoside under CARS-PLS model

References 35

[1]	陈丽梅, 魏文萍, 王慧, 等. 中药覆盆子炮制研究进展[J]. 中华中医药学刊, 2024, 42(12): 138-141.
[2]	ISPIRYAN A, ATKOCIUNIENE V, MAKSTUTIENE N, et al. Correlation between antimicrobial activity values and total phenolic content/antioxidant activity in Rubus idaeus L[J]. Plants, 2024, 13(4): 504.
[3]	GARJONYTE R, BUDIENE J, LABANAUSKAS L, et al. In vitro antioxidant and prooxidant activities of red raspberry (Rubus idaeus L.) stem extracts[J]. Molecules, 2022, 27(13): 4073.
[4]	RAAL A, VAHTRA A, KOSHOVYI O, et al. Polyphenolic compounds in the stems of raspberry (Rubus idaeus) growing wild and cultivated[J]. Molecules, 2024, 29(21): 5016.
[5]	ZHANG X B, ZHAO Y P, GUO L P, et al. Differences in chemical constituents of Artemisia annua L from different geographical regions in China[J]. PLoS One, 2017, 12(9): e0183047.
[6]	PIAO X M, HUO Y, KANG J P, et al. Diversity of ginsenoside profiles produced by various processing technologies[J]. Molecules, 2020, 25(19): 4390.
[7]	YANG R W, HAN B, WANG B, et al. Insights into the potential quality markers of Rubus chingii Hu fruit at different growth stages[J]. Food Research International, 2025, 201: 115552.
[8]	LI Q, QI L M, ZHAO K, et al. Integrative quantitative and qualitative analysis for the quality evaluation and monitoring of Danshen medicines from different sources using HPLC-DAD and NIR combined with chemometrics[J]. Frontiers in Plant Science, 2022, 13: 932855.
[9]	MA H L, ZHAO Y, HE W X, et al. Quantitative analysis of three ingredients in Salvia miltiorrhiza by near infrared spectroscopy combined with hybrid variable selection strategy[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2024, 315: 124273.
[10]	MIAO X X, MIAO Y, LIU Y, et al. Measurement of nitrogen content in rice plant using near infrared spectroscopy combined with different PLS algorithms[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2023, 284: 121733.
[11]	LI G Y, LI J Q, LIU H G, et al. Geographic traceability of Gastrodia elata Blum based on combination of NIRS and chemometrics[J]. Food Chemistry, 2025, 464: 141529.
[12]	PIELORZ S, FECKA I, BERNACKA K, et al. Quantitative determination of polyphenols and flavonoids in Cistus×incanus on the basis of IR, NIR and Raman spectra[J]. Molecules, 2022, 28(1): 161.
[13]	ZHANG S J, GONG X C, QU H B. Near-infrared spectroscopy and HPLC combined with chemometrics for comprehensive evaluation of six organic acids in Ginkgo biloba leaf extract[J]. Journal of Pharmacy and Pharmacology, 2022, 74(7): 1040-1050.
[14]	ZHU Y W, CHEN X Y, WANG S M, et al. Simultaneous measurement of contents of liquirtin and glycyrrhizic acid in liquorice based on near infrared spectroscopy[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2018, 196: 209-214.
[15]	LI L L, LI L, GOU G Y, et al. A nondestructive detection method for the muti-quality attributes of oats using near-infrared spectroscopy[J]. Foods, 2024, 13(22): 3560.
[16]	XU X F, NIE L X, PAN L L, et al. Quantitative analysis of Panax ginseng by FT-NIR spectroscopy[J]. Journal of Analytical Methods in Chemistry, 2014, 2014: 741571.
[17]	YANG Z Y, CHENG Z, SU P Y, et al. A model for the detection of β-glucan content in oat grain based on near infrared spectroscopy[J]. Journal of Food Composition and Analysis, 2024, 129: 106105.
[18]	庄伟岳. HPLC法同时测定覆盆子不同部位中5种有效成分含量[J]. 中国民族民间医药, 2024, 33(22): 16-20.
[19]	XIE C Q, XU N, SHAO Y N, et al. Using FT-NIR spectroscopy technique to determine arginine content in fermented Cordyceps sinensis mycelium[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2015, 149: 971-977.
[20]	ZHANG J B, LI Y, WANG B, et al. Rapid evaluation of Radix Paeoniae Alba and its processed products by near-infrared spectroscopy combined with multivariate algorithms[J]. Analytical and Bioanalytical Chemistry, 2023, 415(9): 1719-1732.
[21]	HARUNA S A, LI H H, WEI W Y, et al. Simultaneous quantification of total flavonoids and phenolic content in raw peanut seeds via NIR spectroscopy coupled with integrated algorithms[J]. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2023, 285: 121854.
[22]	SHI H T, YU P Q. Using molecular spectroscopic techniques (NIR and ATR-FT/MIR) coupling with various chemometrics to test possibility to reveal chemical and molecular response of cool-season adapted wheat grain to ergot alkaloids[J]. Toxins, 2023, 15(2): 151.
[23]	PING F J, YANG J H, ZHOU X J, et al. Quality assessment and ripeness prediction of table grapes using visible-near-infrared spectroscopy[J]. Foods, 2023, 12(12): 2364.
[24]	YE S T, WENG H Y, XIANG L R, et al. Synchronously predicting tea polyphenol and epigallocatechin gallate in tea leaves using Fourier transform-near-infrared spectroscopy and machine learning[J]. Molecules, 2023, 28(14): 5379.
[25]	潘萍, 陈梦婷, 李翱翔, 等. 覆盆子化学成分、药理作用与临床运用研究进展[J]. 中国药师, 2024, 27(1): 155-170.
[26]	彭彬倩, 沈福苗. 近红外光谱技术在食品检测中的应用研究进展[J]. 食品安全导刊, 2025(5): 181-184, 189.
[27]	NAZARLOO A S, SHARABIANI V R, GILANDEH Y A, et al. Evaluation of different models for non-destructive detection of tomato pesticide residues based on near-infrared spectroscopy[J]. Sensors, 2021, 21(9): 3032.
[28]	MANCINI M, MAZZONI L, GAGLIARDI F, et al. Application of the non-destructive NIR technique for the evaluation of strawberry fruits quality parameters[J]. Foods, 2020, 9(4): 441.
[29]	HUANG Z X, LIU L, LI G J, et al. Nondestructive determination of diastase activity of honey based on visible and near-infrared spectroscopy[J]. Molecules, 2019, 24(7): 1244.
[30]	ADNAN A, VON HÖRSTEN D, PAWELZIK E, et al. Rapid prediction of moisture content in intact green coffee beans using near infrared spectroscopy[J]. Foods, 2017, 6(5): 38.
[31]	TIAN Z X, TAN Z F, LI Y J, et al. Rapid monitoring of flavonoid content in sweet tea (Lithocarpus litseifolius (Hance) Chun) leaves using NIR spectroscopy[J]. Plant Methods, 2022, 18(1): 44.
[32]	ZHAO X H, PAN X, YAN W H, et al. Visible-NIR hyperspectral classification of grass based on multivariate smooth mapping and extreme active learning approach[J]. Scientific Reports, 2022, 12(1): 9017.
[33]	JI Q G, LI C F, FU X S, et al. Protected geographical indication discrimination of Zhejiang and non-Zhejiang Ophiopogonis japonicus by near-infrared (NIR) spectroscopy combined with chemometrics: the influence of different stoichiometric and spectrogram pretreatment methods[J]. Molecules, 2023, 28(6): 2803.
[34]	LI M X, SHI Y B, ZHANG J B, et al. Rapid evaluation of Ziziphi Spinosae Semen and its adulterants based on the combination of FT-NIR and multivariate algorithms[J]. Food Chemistry: X, 2023, 20: 101022.
[35]	BAI Z J, XIE M D, HU B F, et al. Estimation of soil organic carbon using vis-NIR spectral data and spectral feature bands selection in southern Xinjiang, China[J]. Sensors, 2022, 22(16): 6124.