In order to overcome the deficiency of the traditional calculation meth od of the multifractal, a mathematical morphology based on multifractal estimati on technique is proposed in this study. Signals acquired from different types of gear condition, i.e. normal condition, gear tooth wear fault and gear tooth broken fault, are employed to evaluate this new method. Results have revealed that the multifractal dimensions calculated by mathematical morphology are very sta ble and have a superior capacity to distinguish the different gear states. Otherwise, mathematical morphology operation containes only the simple addition, subtr action, maximum and minimum which can dramatically reduce the computation cost compared to other technique. Thus the proposed method has demonstrated a more fast and effective technique for accurate fault diagnosis of gear.