Spectroscopy and Spectral Analysis, Volume. 41, Issue 12, 3665(2021)
A Method of Terahertz Spectrum Material Identification Based on Wavelet Coefficient Graph
Figures & Tables(6)
Table 1. Summary of experimental sample
View tableView in ArticleTable 1. Summary of experimental sample
类别 测试样品 含吸收峰 乳糖、 葡萄糖、 果糖、 蔗糖、 麦芽糖、 山梨酸钾、 苯甲酸、 苯甲酸钠、 叔丁基氢醌、 滑石粉 无明显吸收峰 红磷、 淀粉、 乙基麦芽酚、 甲基环戊烯醇酮、 溴化钾、 氯化钾、 氯化钙、 硫酸钾、 硫酸钠、 聚乙烯
Table 2. Summary of pure substances recognition results
View tableView in ArticleTable 2. Summary of pure substances recognition results
第一组 第二组 第三组 CNN a 100% 98.13% 97.19% b 100% 100% 100% SVM a 32.22% 16.67% 20.27% b 76.39% 58.33% 62.50% MLP a 94.44% 90.27% 81.94% b 97.22% 94.44% 86.11%
Table 3. The recognition rate summary of the two methods
View tableView in ArticleTable 3. The recognition rate summary of the two methods
小波寻峰识别 本文 识别率 平均相似度 识别率 平均相似度 麦芽糖 90% 61.17% 100% 97.14% 果糖 35% 38.51% 92.86% 89.51% 乳糖 100% 95.07% 92.86% 97.33% 葡萄糖 100% 71.69% 100% 97.60% 蔗糖 25% 38.88% 100% 91.69% 淀粉 - - 100% 89.44%
Tools
Get Citation
Copy Citation Text
Yan-ling CHEN, Liang-lun CHENG, Heng WU, Li-min XU, Wei-jian HE, Feng LI. A Method of Terahertz Spectrum Material Identification Based on Wavelet Coefficient Graph[J]. Spectroscopy and Spectral Analysis, 2021, 41(12): 3665
Paper Information
Category: Research Articles
Received: Nov. 16, 2020
Accepted: --
Published Online: Dec. 17, 2021
The Author Email: Yan-ling CHEN (jkjkre@qq.com)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20