Based on the bus IC card data, this paper combines the land types around the bus station and the passenger space-time transfer sequence to construct a Hidden Markov model to identify the travel purpose of the passenger, and then realizes the extraction of a complete passenger bus travel chain. Using Shijiazhuang"s bus IC card data for a working day to implement the model, and comparing it with the bus passenger travel survey data of the same date, the results show that the model is more in line with the actual survey results. The bus passenger travel chain is dominated by commuting travel chains , With a total proportion of 81.87%, and commuting activities have a strong timeliness. The model in this paper provides a certain basis for in-depth research on public transit IC card data.