Spectroscopy and Spectral Analysis, Volume. 42, Issue 2, 440(2022)
A Wavelength Selection Method Combining Direct Orthogonal Signal Correction and Monte Carlo
Figures & Tables(7)
Fig. 1. The comparison of PLS regression coefficients and DOSC weights, where
Fig. 2. (a) The number of wavelength points selected in each iteration; (b) RMSECV in each iteration; (c) Comparison of RMSEP before and after wavelength selection
Fig. 3. (a) The number of wavelength points selected in each iteration; (b) RMSECV in each iteration;(c) Comparison of RMSEP before and after wavelength selection
Table 1. Comparison of prediction ability of different models in corn dataset
View tableView in ArticleTable 1. Comparison of prediction ability of different models in corn dataset
模型 波长点个数 测试集相关系数Rp RMSEP PLS 700 0.828 2 0.109 8 MCUVE 296 0.862 4 0.099 2 GA 32 0.865 3 0.098 2 CARS 17 0.916 2 0.078 5 MC-DOSC 15 0.931 4 0.071 3
Table 2. Comparison of prediction ability of different models in gasoline data sets
View tableView in ArticleTable 2. Comparison of prediction ability of different models in gasoline data sets
模型 波长点个数 测试集相关系数Rp RMSEP PLS 301 0.987 5 0.255 5 MCUVE 68 0.990 3 0.225 2 GA 43 0.991 8 0.206 6 CARS 44 0.991 1 0.215 2 MC-DOSC 31 0.993 9 0.178 8
Tools
Get Citation
Copy Citation Text
Lin-jiang XIE, Ming-jian HONG, Zhi-rong YU. A Wavelength Selection Method Combining Direct Orthogonal Signal Correction and Monte Carlo[J]. Spectroscopy and Spectral Analysis, 2022, 42(2): 440
Paper Information
Category: Research Articles
Received: Jan. 7, 2021
Accepted: Apr. 11, 2021
Published Online: Apr. 2, 2022
The Author Email: XIE Lin-jiang (584283862@qq.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20