光谱滤波法提高激光诱导击穿光谱对蔬菜中元素Pb的检测精度

发布时间：2018-05-15 20:13

本文选题：蔬菜 + 铅　；参考：《分析化学》2017年08期

【摘要】：激光诱导击穿光谱(Laser induced breakdown spectroscopy,LIBS)原始光谱中包含较多噪声信号,为探究不同滤波方法对LIBS光谱预处理的影响,本研究以实验室Pb污染处理的蔬菜为研究对象,采集波长范围在400.45~410.98 nm的LIBS谱线信息,分别利用相邻平均(Adjacent averaging)、Savitzky-Golay(S-G)滤波器、快速傅里叶变换(Fast Fourier transformation,FFT)对采集的LIBS光谱进行平滑、去噪,并结合偏最小二乘法(PLS)定量分析模型对光谱处理效果进行评价。结果表明,S-G平滑效果最优,当S-G滤波器窗口宽度为15,拟合阶次为3时,PLS定量模型效果最佳,其验证集均方根误差(RMSEP)为0.26、平均相对误差(ARE)为3.7%。结果表明,选择适合的滤波方法有助于提高LIBS光谱质量以及检测模型的精度。
[Abstract]:Laser induced breakdown spectroscopy (LIBS) contains more noise signals in the original spectrum. In order to explore the effect of different filtering methods on LIBS spectrum pretreatment, the vegetables contaminated by Pb in laboratory were used as the research object in this study. LIBS line information with wavelength range of 400.45 ~ 410.98 nm is collected. The LIBS spectrum is smoothed and de-noised by using adjacent average Adjacent averaginger Savitzky-Golayay S-G filter and Fast Fourier transform (FFT). The spectral processing effect was evaluated with the partial least square (PLS) quantitative analysis model. The results show that the S-G smoothing effect is the best, when the window width of S-G filter is 15, the fitting order is 3, the RMSEPs of the verification set is 0.26, and the average relative error is 3.7. The results show that the selection of suitable filtering method is helpful to improve the LIBS spectral quality and the accuracy of the detection model.
【作者单位】：江西省高校生物光电及应用重点实验室;江西省果蔬采后处理关键技术及质量安全协同创新中心;
【基金】：国家自然科学基金(Nos.31460419、31560482) 江西省科技支撑计划项目(No.20151BBG70063) 江西省重大科技项目自然科学基金资助项目(No.20143ACB21013) 江西省远航工程计划项目(No.20140142)资助~~
【分类号】：O657.3;TS255.7

【相似文献】