Study on Identification Methods of Pericarpium Citri Reticulatae Based on Time-Gated Raman Spectroscopy and Support Vector Machine
FENG Hao-heng1, 2, 3, 4, LI Ke-qing5, 6, NIU Yuan-yuan2, 4, SUN Qing-di2, 4, XU Jin-chang2, 4, FANG Guang-you1, 2, 3, 4, CHEN Jian5, 6, HU Min7, 8, 9*, WANG Zhen-you1, 2, 3, 4*
1. Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
2. GBA Branch of Aerospace Information Research Institute, Chinese Academy of Sciences, Guangzhou 510700, China
3. School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, China
4. Guangdong Provincial Key Laboratory of Terahertz Quantum Electromagnetics, Guangzhou 510700, China
5. School of Chemistry, Sun Yat-sen University, Guangzhou 510006, China
6. Instrumentation Analysis & Research Center, Sun Yat-sen University, Guangzhou 510275, China
7. Terahertz Research Center, School of Electronic Science and Engineering, University of Electronic Science and Technology of China, Chengdu 610054, China
8. Key Laboratory of Terahertz Technology, Ministry of Education, Chengdu 610054, China
9. Tian Fu Jiang Xi Laboratory, Chengdu 641419, China
Abstract：The efficacy and price of Pericarpium Citri Reticulatae (PCR) are significantly influenced by its origin and harvest year, necessitating a rapidand effective identification method.Traditional chemical analysis techniques, though accurate, are complex, expensive, and time-consuming. Raman spectroscopy, with its high specificity and non-destructive detection capabilities, is a promising method for rapid detection. However, the strong fluorescence background of PCR limits the application of conventional Raman spectroscopy. To address this issue, this study combines time-gated Raman (TG-Raman) spectroscopy with support vector machine (SVM) classification models, proposing an efficient and non-destructive method for identifying PCR. The study selected six groups of PCR samples from different origins and ages in Xinhui District, Jiangmen City, Guangdong Province, as well as one crafted PCR sample, and compared the 532 nm, 1 064 nm continuous wave, and 532 nm TG-Raman spectra. The experimental results demonstrate that TG-Raman spectroscopy effectively eliminates fluorescence interference, thereby significantly enhancing the signal-to-noise ratio of the Raman signals and facilitating the extraction of more characteristic Raman peaks for chemical components.The key Raman peaks of PCR were observed at 856, 1 084, 1 112, 1 264, 1 300, 1 340, 1 456, 1 607, and 2 935 cm-1. Spectral analysis revealed that the main chemical components of PCR include pectin, cellulose, fatty acids, and flavonoids. Among these, the flavonoid characteristic peak at 1 607 cm-1 exhibited significant intensity variation, making it a key marker for distinguishing PCR from different origins and ages. Based on the extracted spectral features, the study constructed SVM classification models using various kernel functions, optimizing model parameters, and found that the radial basis function (RBF) kernel performed best. After training and testing on different PCR samples, the highest classification accuracy reached 96.43%, fully demonstrating the excellent classification performance of the combination of TG-Raman spectroscopy and SVM. The study indicates that this method is efficient and non-destructive, enabling accurate identification of PCR samples in a short time, with broad application potential in origin tracing and age identification. In conclusion, this study -presents a novel non-destructive detection technique for PCR and other medicinal materials, offering significant advantages in addressing the time-consuming and high-cost issues associated with traditional chemical analysis methods. This technology offers robust technical support for the quality control, authenticity verification, and traceability of medicinal materials, with wide-ranging application prospects.
冯浩恒，李可清，牛园园，孙青荻，徐金畅，方广有，陈 建，胡 旻，王振友. 基于时间门控拉曼光谱和支持向量机的陈皮鉴别方法研究[J]. 光谱学与光谱分析, 2025, 45(09): 2557-2562.
FENG Hao-heng, LI Ke-qing, NIU Yuan-yuan, SUN Qing-di, XU Jin-chang, FANG Guang-you, CHEN Jian, HU Min, WANG Zhen-you. Study on Identification Methods of Pericarpium Citri Reticulatae Based on Time-Gated Raman Spectroscopy and Support Vector Machine. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(09): 2557-2562.
[1] Wang Q, Qiu Z, Chen Y, et al. Critical Reviews in Food Science and Nutrition, 2024, 64(28): 10332.
[2] Yu X, Sun S, Guo Y, et al. Journal of Ethnopharmacology, 2018, 220: 265.
[3] Fu M, Xu Y, Chen Y, et al. Food Chemistry, 2017, 230: 649.
[4] Li Y, Zhao W, Qian M, et al. Food Chemistry, 2024, 442: 138531.
[5] Tang L J, Li X K, Huang Y, et al. Food Chemistry: X, 2024, 23: 101759.
[6] Liu Y, Pu H, Li Q, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2023, 286: 122035.
[7] Chen D D, Xie X F, Ao H, et al. Journal of the Chinese medical Association, 2017, 80(5): 288.
[8] Xu C X, Song P, Yu Z, et al. Analyst, 2024, 149(1): 46.
[9] Holtum T, Reichenauer T, Kumar V, et al. Journal of Raman Spectroscopy, 2023, 54(3): 288.
[10] Lipiaäinen T, Pessi J, Movahedi P, et al. Analytical Chemistry, 2018, 90(7): 4832.
[11] Itkonen J, Ghemtio L, Pellegrino D, et al. Pharmaceutics, 2022, 14(8): 1639.
[12] Inoue T, Yoshinaga A, Takabe K, et al. Phytochemical Analysis, 2015, 26(2): 105.
[13] Mazurek S, Włodarczyk M, Pielorz S, et al. Molecules, 2022, 27(12): 3954.
[14] Pompeu D R, Larondelle Y, Rogez H, et al. Biotechnologie Agronomie Societe et Environnment, 2018, 22: 13.
[15] Synytsya A, ( Cˇ )opíková J, Matějka P, et al. Carbohydrate Polymers, 2003, 54(1): 97.
[16] Szymańska-Chargot M, Chylińska M, Pieczywek P M, et al. Planta, 2016, 243: 935.
[17] Feng X, Zhang Q, Zhu Z. Food Science and Technology Research, 2013, 19(6): 1077.
[18] Schulz H, Baranska M. Vibrational Spectroscopy, 2007, 43(1): 13.
[19] Wiercigroch E, Szafraniec E, Czamara K, et al. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2017, 185: 317.
[20] Yang Y, Wang X, Zhao C, et al. Journal of Food Science, 2017, 82(12): 2840.
[21] Nekvapil F, Brezestean I, Barchewitz D, et al. Food Chemistry, 2018, 242: 560.
[22] Fu Y, Wang C, Gao Z, et al. Food Chemistry, 2025, 464: 141631.
