Near infrared (NIR) spectroscopy coupled with one-class support vector machine (OC-SVM) were used to rapidly and accurately monitor physical and chemical changes in solid-state fermentation (SSF) of crop straws without the need for chemical analysis. Raw spectra of fermented samples were acquired with wavelength range of 10000~4000 cm-1. Then the top seven PCs as input vectors were extracted by principal component analysis (PCA). OC-SVM algorithm was implemented to develop identification model, and some parameters of OC-SVM model were optimized by cross-validation in calibrating model. Experimental results showed that OC-SVM model revealed its incomparable superiority than SVM model in handling imbalance training sets under the same condition. The discrimination rate of OC-SVM model was 85% in the validation set when the ratio of samples from target class to those from non-target class was one to eight in the training set.