Discrimination of wood biological decay by NIR and partial least squares discriminant analysis (PLS-DA)

Extensive research has demonstrated that near infrared spectroscopy (NIR) and partial least squares discriminant analysis (PLS-DA) can be used to rapidly discriminate or detect a wide variety of food, medicine and agricultural products. The use of NIR coupled with PLS-DA to detect wood biological decay was investigated in the present paper. The results showed that the correlation between the predicted category variable of calibration and validation and the measured category variable is significant with a correlation coefficient (r) over 0.94 and low SEC and SEP (< 0.17); the discriminant accuracy for the non-decay, whiterot and brown-rot decay samples are 100% (deviation < 0.5) by the PLS-DA model based on the test set of samples; the discriminant accuracy by PLS-DA model is better than that by SIMCA model. It's suggested that NIR spectroscopy coupled with PLS-DA could be used to rapidly detect wood biological decay.