Application of Near-Infrared Hyperspectral Imaging with Variable Selection Methods to Determine and Visualize Caffeine Content of Coffee Beans

详细信息    查看全文

• 作者：Chu Zhang ; Hao Jiang ; Fei Liu ; Yong He
• 关键词：Coffee ; Random frog ; Prediction map ; Partial least squares regression
• 刊名：Food and Bioprocess Technology
• 出版年：2017
• 出版时间：January 2017
• 年：2017
• 卷：10
• 期：1
• 页码：213-221
• 全文大小：
• 刊物类别：Chemistry and Materials Science
• 刊物主题：Food Science; Chemistry/Food Science, general; Agriculture; Biotechnology;
• 出版者：Springer US
• ISSN：1935-5149
• 卷排序：10

文摘

Hyperspectral imaging covering the spectral range of 874–1734 nm was used to determine caffeine content of coffee beans. Spectral data of 958.24–1628.89 nm were extracted and preprocessed. Partial least squares regression (PLSR) model on the preprocessed full spectra obtained good performance with coefficient of determination of prediction (R2_p) of 0.843 and root mean square error of prediction (RMSEP) of 131.904 μg/g. In addition, 10 variable selection methods were applied to select the best optimal wavelengths. The PLSR models on the different optimal wavelengths obtained satisfactory results. The PLSR model on the wavelengths selected by random frog (RF) performed the best, with R2_p of 0.878 and RMSEP of 116.327 μg/g. The RF wavelength selection combined with the PLSR model also achieved satisfactory visualization of caffeine content between different coffee beans. The overall results indicated that optimal wavelength selection was an efficient method for spectral data preprocessing, and hyperspectral imaging was illustrated as a potential technique for real-time online determination for caffeine content of coffee beans.