Spectroscopy and Spectral Analysis, Volume. 42, Issue 9, 2830(2022)
Support Vector Machine Optimized by Near-Infrared Spectroscopic Technique Combined With Grey Wolf Optimizer Algorithm to Realize Rapid Identification of Tobacco Origin
Figures & Tables(6)
Fig. 2. CARS variable screening results
(a): Number of wavelengths changed with sampling runs; (b): RMSECV changed with sampling runs; (c): Variable coefficient changed with sampling runs
Fig. 4. Confusion matrices of different classification models
(a): RF-PSO-SVM; (b): RF-GA-SVM; (c): RF-GWO-SVM
Table 1. Training and testing sample sizes of tobacco from different regions
View tableView in ArticleTable 1. Training and testing sample sizes of tobacco from different regions
分类编号 产地 样品编号 校正集 验证集 总数 1 福建 98 40 138 2 广西 52 8 60 3 贵州 68 26 94 4 河南 79 33 112 5 湖北 55 44 99 6 湖南 96 16 112 7 四川 86 17 103 8 云南 83 23 106
Table 2. Classification effects of different models
View tableView in ArticleTable 2. Classification effects of different models
建模方法 波长选择方法 变量数 最佳C值 最佳σ值 运行时间/min 校正集正确率/% 验证集正确率/% PSO-SVM CARS 141 37.07 0.01 47 94.00 87.92 RF 534 33.00 0.01 178 99.68 95.17 GA-SVM CARS 141 4.61 0.37 39 100.00 91.30 RF 534 15.58 0.03 153 100.00 94.20 GWO-SVM CARS 141 6.07 0.11 10 100.00 93.24 RF 534 34.48 0.10 22 100.00 96.62
Tools
Get Citation
Copy Citation Text
Ying-rui GENG, Huan-chao SHEN, Hong-fei NI, Yong CHEN, Xue-song LIU. Support Vector Machine Optimized by Near-Infrared Spectroscopic Technique Combined With Grey Wolf Optimizer Algorithm to Realize Rapid Identification of Tobacco Origin[J]. Spectroscopy and Spectral Analysis, 2022, 42(9): 2830
Paper Information
Category: Research Articles
Received: Aug. 17, 2021
Accepted: Nov. 16, 2021
Published Online: Nov. 17, 2022
The Author Email: GENG Ying-rui (22019084@zju.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20