In pharmacological experiments, a popular method to discover the effects of psychotherapeutic drugs is to monitor behaviors of laboratory mice subjected to drugs by cameras. Automating behavior analysis of laboratory mice saves both time and human labor. In this study, we focus on automated action recognition of laboratory mice from short video clips in which only one action is performed. A two-stage recognition method is designed to address the problem. In the first stage, still actions such as sleeping are separated from other action classes based on the amount of the motion area. Remaining action classes are discriminated by the second stage in which we project 3D action volume onto 2D images by encoding temporal variations of each pixel using discrete wavelet transform (DWT). Resulting images are modeled and classified by hidden Markov models in maximum likelihood sense. We test the proposed action recognition method on a publicly available mice action dataset and achieve promising recognition rates. In addition, we compare our method to well-known studies in the literature.