PLS-DA法判别分析木材生物腐朽的研究

Discrimination of Wood Biological Decay by NIR and Partial Least Squares Discriminant Analysis (PLS-DA)

利用近红外光谱结合PLS-DA判别分析方法可用于食品、药品和农产品等的快速识别或检测,因此,研究利用近红外光谱结合PLS-DA方法来检测木材的生物腐朽。研究结果表明:应用近红外光谱结合PLS-DA方法对培训集样本建立的判别模型,其校正及验证结果与实际分类变量的相关系数均超过0.94,SEC和SEP都低于0.17;利用模型对未参与建模的样本进行检测,发现该模型对未腐朽、白腐和褐腐三种类型样本的判别准确率均为100%(偏差均小于0.5);与SIMCA法相比,PLS-DA法对木材生物腐朽样本的判别准确率更高,说明应用近红外光谱结合PLS-DA方法能快速地检测到木材的生物腐朽,并能准确地判别出木材的生物腐朽类型。

Extensive research has demonstrated that near infrared spectroscopy （NIR） and partial least squares discriminant anal- ysis （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, white rot 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.