The invention discloses a gait recognition method based on regional entropy characteristics. The method comprises three parts of preprocessing, feature extraction and classification recognition. The method comprises the following specific steps: pre-processing for obtaining a moving target contour sequence with a standard uniform size, comprising the following steps: 1-1. extracting a moving target contour sequence; 1-2. image standardization; the feature extraction used for obtaining gait feature parameters with good characterization, comprising the following steps: 2-1, gait period extraction; 2-2, extracting regional image entropy characteristics; classification and identification by adopting a nearest neighbor criterion. The extracted regional image entropy features have good characterization, a complex image processing process is not needed, the extraction mode is simple, the steps of dimensionality reduction and data normalization are avoided, the system complexity is reduced, and the higher recognition rate of gait identity authentication is obtained.