Savitzky, A. and Golay, M.J.E. (1964) Smoothing and Differentiation of Data by Simplified Least Squares Procedures, Analytical Chemistry, 36, 1627-1637.
https://doi.org/10.1021/ac60214a047

has been cited by the following article:

• TITLE: Rapid Analysis of Soil Copper Content for the Pearl River Delta Based on Near-Infrared Spectroscopy Combined with SG-PLS

  AUTHORS: Guangwei Li, Jing Zhang, Tao Pan, Jing Yu, Lijun Yao

  KEYWORDS: Soil, Copper, Near-Infrared Spectroscopy, Savitzky-Golay Smoothing, Partial Least Squares Regression

  JOURNAL NAME: American Journal of Analytical Chemistry, Vol.9 No.12, December 27, 2018

  ABSTRACT: Using near-infrared (NIR) spectroscopy combined with an optimal method for Savitzky-Golay (SG) smoothing and partial least squares (PLS) regression, a rapid analysis method was established for copper content in the beach reclamation soil samples from Pearl River Delta in China. A framework with calibration, prediction and validation was established by considering randomness and stability. The parameters were optimized according to the comprehensive index (SEP+) to produce modeling stability. The validation results show that, based on the SG-PLS model in long-NIR region (1100 - 2498 nm) with first-order derivative, fifth degree polynomial, seven smoothing points and six PLS factors, the corresponding root mean square error (SEP), correlation coefficient of prediction (RP) and average relative error (ARE) were 0.31 mg·kg-1, 0.924 and 4.5%, respectively. The result indicates high prediction accuracy. The relevant parameter selection can also provide a reference for designing small and dedicated spectrometer.