Wine component tracing method based on near infrared spectrum fusion machine learning

An intelligent wine detection and traceability method based on infrared spec-troscopy machine learning is proposed, in order to meet the needs of online rapid nondestructive testing wine. On basis extracting spectrum wine, principal component analysis (PCA) – support vector (SVM) model was modified by chemometrics. A total 300 grape samples were collected from six production areas. The composition analyzed ultra performance liquid chromatography-quadrupole time-of-flight mass spectrometry (UPLC-Q-TOF-MS). According experimental results, indole, sulfacetamide caffeine selected as characteristics different origins. Near spectral wavelengths compressed between 900 2,500 nm. ranges 1,000 nm ~ 1,400 1,500 1800 for PCA key extracted. unsupervised SVM used classify identify wavelengths. results show that algorithm has higher classification accuracy than traditional PCA-LDA, other algorithms. improved 98.3 99.75%. PCA-SVM can achieve fast loss-less source tracing