| 초록 |
This study was performed to prediction of ingredient variable of Hanji using near infrared spectroscopy (NIR) and partial last square discriminant analysis (PLS-DA) method. In order to perform this, 12 of Hanji including different kinds of paper mulberry, cooking agent and fiber dispersal agent were manufactured. Then, the samples were analyzed by Fourier Transform near Infrared spectroscopy (FT-NIR) at 12,500 to 4,000 cm−1. The models classified by cross validation, accuracy, sensitivity and specificity were calculated to evaluate the classification performance of PLS-DA models. |
