Abstract: A process of extracting the temporal and spatial distribution characteristics of urban residents by using taxi trajectory data is introduced, including: using the method of mathematical statistics to analyze the time-based characteristics of taxi boarding and alighting events; A density clustering algorithm integrating kernel density estimation (KDE) and point of interest (POI) classification is proposed, which realizes the mining of taxi loading and unloading hot spots and the discovery of the relationship between residents' travel activity law and urban functional areas. The research shows that the trajectory characteristics of residents show obvious differences between “work-rest” days and different periods, and this difference is closely related to the distribution of urban functional areas.