Abstract:A new method based on near infrared spectroscopy technology was established to detect peanut oil adulteration quickly. Four kinds of adulteration samples of soybean oil, rapeseed oil, palm oil and blend oil and five samples of pure peanut oil were collected for detection. Near Infrared spectrums of the samples were normalized as the input variables to SVM. The kernel function of SVM was RBF. The best combination of penalty parameter c and RBF kernel parameter γ were selected by gird search and K –fold cross validation (c=1024,γ=1). The recognition ratio and prediction ratio were 100%. The experiments showed that the fast detection technology based on NIR and SVM was practicable for identifying peanut oil adulteration.