The inner ring erosion and outer indentation are typical faults of rolling bearing. In order to diagnose these faults rapidly and accurately, a novel diagnosis method of rolling bearing was proposed based on the energy characteristics of PF (Product Function) component and support vector machine (Support Vector Machine, SVM) by the vibration signal of local mean decomposition(Local mean decomposition, LMD). The collected vibration signals were decomposed into several PF components by the local mean decomposition, the calculated energy feature of the PF component were inputted to the support vector machine to identify the type of rolling bearing faults. The results show that the method has a high diagnosis and recognition rate for the typical faults of rolling bearing.