基于PCA-SVM的合成色素SERS信号判别
SERS Signal Discrimination of Synthetic Pigment Based on PCA-SVM

DOI: 10.12677/CSA.2020.103045, PP. 437-444

Keywords: SERS，主成分分析，支持向量机，色素
SERS, PCA, SVM, Pigment

Full-Text   Cite this paper   Add to My Lib

Abstract:

表面增强拉曼光谱(SERS)是一种新型的物质检测技术，有快速、高效、低损耗率等优点。在进行成分分析时，具有相似结构的待分析物的SERS光谱会出现堆叠的情况，难以采用常规方法区分这类物质。本文基于主成分分析法(PCA)和支持向量机(SVM)相结合的模型对光谱进行分类预测，以同类型的可食用人工合成色素为例，运用SERS和PCA-SVM模型，验证了该分类模型的有效性。结果表明，该方法对不同色素预测的准确度高达98%，且所呈现的结果基本与预期结果相同，具有良好的分类效果。为相似结构物质的SERS信号处理提供依据。
Surface enhanced Raman spectroscopy (SERS) is a novel substance detection technology, which has the advantages of fast, high efficiency, and low loss rate. In the analysis of components, it is difficult to distinguish them by conventional means because of the high overlapping of SERS spectra of analytes with similar structures. In this paper, based on the combination of principal component analysis (PCA) and support vector machine (SVM) model to classify and predict the spectrum, taking the edible synthetic pigment of the same type as an example, using SERS and PCA-SVM model to verify the effectiveness of the classification model, the results show that the accuracy of this method is as high as 98%, and the results are basically the same as expected. It provides a basis for SERS signal processing of similar structure substances.

References

[1]	Han, C.Q., Yao, Y. and Wang, W. (2017) Rapid and Sensitive Detection of Sodium Saccharin in Soft Drinks by Silver Nanorod Array SERS Substrates. Sensors and Actuators B-Chemical, 251, 272-279. https://doi.org/10.1016/j.snb.2017.05.051
[2]	Yan, B., Li, B. and Wen, Z.N. (2015) Label-Free Blood Serum Detection by Using Surface-Enhanced Raman Spectroscopy and Support Vector Machine for the Preoperative Diagnosis of Parotid Gland Tumors. BMC Cancer, 15, 650. https://doi.org/10.1186/s12885-015-1653-7
[3]	王昕, 何坚, 范贤光.基于表面增强拉曼光谱的合成色素专利蓝V的快速检测[J]. 光谱学与光谱分析, 2019, 39(10): 3205-3209.
[4]	Ma, C.-H., Zhang, J. and Hong, Y.-C. (2015) Determination of Carbendazim in Tea Using Surface Enhanced Raman Spectroscopy. Chinese Chemical Letters, 26, 1455-1459. https://doi.org/10.1016/j.cclet.2015.10.015
[5]	Aditya, K. and Venugopal, S. (2019) Paper Swab Based SERS Detection of Non-Permitted Colourants from Dals and Vegetables Using a Portable Spectrometer. Analytica Chimica Acta, 1090, 106-113. https://doi.org/10.1016/j.aca.2019.08.073
[6]	牛宏侠, 张肇鑫, 宁正. 基于小波变换的阈值自适应寻优去噪方法[J/OL]. 传感器与微系统, 2020(2): 33-36.
[7]	毛烨, 陈亮. 强噪声条件下激光雷达强度图像降噪方法研究[J]. 激光杂志, 2019, 40(11): 80-83.
[8]	杜馨, 孙晓荣, 刘翠玲. 拉曼光谱结合偏最小二乘法对食用油品质快速检测研究[J]. 中国酿造, 2019, 38(12): 171-174.
[9]	滕宇. 偏最小二乘法在光谱分析中的应用[J]. 集成电路应用, 2020, 37(1): 16-17.
[10]	公滨南. BP神经网络算法在卫星遥感影像地物分类中的应用[D]: [硕士学位论文]. 哈尔滨: 东北农业大学, 2019.
[11]	全青青. 基于BP神经网络的落叶松人工林树种分类[J]. 信息技术与信息化, 2019(9): 43-45.
[12]	Lin, Z.S. and He, L. (2019) Recent Advance in SERS Techniques for Food Safety and Quality Analysis: A Brief Review. Current Opinion in Food Science, 28. https://doi.org/10.1016/j.cofs.2019.10.001
[13]	刘志强, 吕学, 张利. 基于多分类GA-SVM的高速公路AID模型[J]. 系统工程理论与实践, 2013, 33(8): 2110-2115.
[14]	高金贺, 黄伟玲, 蒋浩鹏. 城市交通碳排放预测的多模型对比分析[J/OL]. 重庆交通大学学报(自然科学版): 1-7[2019-11-14].
[15]	郭云香, 陈龙, 李晓瑾. 基于NIRS技术和PCA-SVM算法快速鉴别国产和进口啤酒花[J]. 药学实践杂志, 2019, 37(4): 322-331.
[16]	陈勇, 李鹏, 张忠军. 基于PCA-GA-LSSVM的输电线路覆冰负荷在线预测模型[J]. 电力系统保护与控制, 2019, 47(10): 110-119.
[17]	刘翠翠, 杨涛, 马京晶. 基于PCA-SVM的麦冬叶部病害识别系统[J]. 中国农机化学报, 2019, 40(8): 132-136.
[18]	段凌风, 徐璐, 王李冬. 基于PCA-BP神经网络的合成色素SERS信号判别[J]. 电脑知识与技术, 2017, 13(31): 196-198+204.
[19]	李海洋, 刘胜. 纺织品近红外光谱定性分析的一种新方法[J]. 光谱学与光谱分析, 2019, 39(7): 2142-2146.
[20]	Hou, X.W., Wang, G.L. and Su, G.Q. (2019) Rapid Identification of Edible Oil Species Using Supervised Support Vector Machine Based on Low-Field Nuclear Magnetic Resonance Relaxation Features. Food Chemistry, 280, 139-145.

Full-Text