It is a key issue for a service robot to deal with different kinds of dynamic obstacles in indoor environments. An approach to track and classify dynamic obstacles in indoor environments is proposed in this paper. A laser range finder (LRF) is used to detect and acquire information of dynamic obstacles. Accordingly dynamic obstacles are grouped into three categories: active obstacles, inactive obstacles and inactive obstacles combined with active obstacles. Three typical obstacles belonging to three categories are analysed based on frequency analysis respectively. According to relative power spectrums, two attributes are put forward to classify three classes of dynamic obstacles. The classification approach could be beneficial to navigation when coping with dynamic obstacles in the indoor environment.