偏最小二乘回归
即时
近红外光谱
校准
分析化学(期刊)
化学
傅里叶变换红外光谱
内容(测量理论)
傅里叶变换
频谱分析仪
光谱学
数学
色谱法
光学
食品科学
统计
量子力学
物理
数学分析
作者
Ying-guo L Uuml,Jie Chen,Xueqin Li,Chun Wang,Lili Wei
出处
期刊:Food Science and Technology Research
[Karger Publishers]
日期:2013-01-01
卷期号:19 (3): 393-398
被引量:5
摘要
A rapid, non-destructive method, based on Fourier transform near-infrared (FT-NIR) spectroscopy for water and oil content determination of instant noodles was presented. A calibration set of 80 samples, a validation set of 40 samples and a prediction set of 10 samples of instant noodles were used. The diffuse reflection spectra of samples were measured by FT-NIR analyzer in the 1000 − 2500 nm spectral range. The optimal models for water and oil content were constructed by comparing different preprocessing method and partial least squares (PLS) factor number. Prediction results indicated good predictive ability of the developed models we set for water and oil content in instant noodles. The correlations between FT-NIR values and chemical values of the two indexes were 0.9912 and 0.9766, respectively. The study provided a rapid method for instant noodles quality control using FT-NIR spectra combine with the method developed in this paper.
科研通智能强力驱动
Strongly Powered by AbleSci AI