Abstract:Aiming at problems that the research of fault classification on bearing is more but the research of fault degree identification is litter, the information of vibration image is abundance but it has not been used fully, the method of fault degree identification of bearing using vibration image is proposed. The original vibration signals are de-noised with EMD- morphology filter, then is converted to bispectrum contour image, after it calculated using gray-level co-occurrence matrix and principal component analysis, the fault degree character parameters are acquired. At last the fault degree is diagnosed by support vector machine. The results of experiments show that this method can diagnose the fault degree of bearing effectively, and it provides a new diagnosis approach for the fault degree of rotating machinery.