Partial Squares-discriminant Analysis Research Articles

다변량 분석과 초분광 이미지를 이용한 고춧가루의 매운맛 등급화

In this study, the spectral image of red pepper powder, which had been prepared in accordance with the standard particle size distribution ratio, was acquired in the short-wave infrared region using a hyperspectral camera. Spectral information was analyzed using multivariate statistical analyses including principal component analysis (PCA) and least partial squares (PLS) analysis. PCA revealed that powders were grouped according to their pungency level, regardless of their particle size distribution (PC1=97%, PC2=2%). The regression coefficient derived in PLS discriminant analysis indicated that 1,201-1,226 nm, 1,387-1,411 nm, and 1,508-1,529 nm are key wavelengths that are affected by the vibration of C-H, O-H, and N-H bonds present in capsaicinoid molecules. Pungency grade was successfully determined, and capsaicinoid content was predicted with high accuracy using PLS analysis of raw data at key wavelength (Rc2=0.9389, Rp2= 0.9261). It was possible to reduce the time required for data calculation and analysis by reducing the amount of spectral data utilized to predict spiciness from 256 to 21 bands. Finally, the distribution of capsaicinoids was mapped visually according to particle size. In conclusion, hyperspectral imaging is a suitable technology for real time, non-destructive monitoring of red pepper powder quality relative to the standard method used during the manufacturing process.

Rapid and non-destructive discrimination of tea varieties by near infrared diffuse reflection spectroscopy coupled with classification and regression trees

The current study attempted to rapidly and non-destructively discriminate the diverse varieties of tea (that is, Biluochun, Longjing, Maojian, Qihong, Tieguanyin, and Yinzhen) via utilizing near infrared (NIR) diffuse reflectance spectroscopy coupled with pattern recognition strategies. Before the recognition analysis, the original NIR spectra were pre-processed by second derivative treatment followed by informative wavenumber interval location. And then, non-linearity detection and outlier diagnosis were performed. When pattern recognition referred, principal component analysis (PCA) was firstly applied to ascertain the discrimination possibility with the NIR spectra. Classification and regression trees (CART), compared with linear discriminant analysis (LDA), and partial squares-discriminant analysis (PLS-DA), was then employed for establishing the discrimination rule. Experimental results showed that the tea quality could be accurately, rapidly, and non-invasively identified via NIR spectroscopy coupled with CART. Key words : Near infrared diffuse reflection spectroscopy, classification and regression trees, and tea variety discrimination.