In the beginning of 2020, the COVID-19 epidemic broke out and spread all over the world in just a few months. COVID-19 spread analysis has attracted considerable research efforts in many areas including the impact of population mobility on the epidemic development. However, most studies do not use real data on population mobility, or choose an overly wide range of objects. This paper studies the ... [Show full abstract] COVID-19 epidemic in Shenzhen from January 26 to February 16, focusing on the impact of population mobility on the epidemic development. Combined with the population mobility data, we propose the Source-SEIR model. We estimated that the basic reproduction number of SARS-CoV-2 is 2.61. The experiment results show that the combination of population mobility data is helpful to the evaluation of the epidemic development, and the restrictions on population mobility in Shenzhen have played a role in curbing the deterioration of COVID-19 epidemic. Without restrictions on population mobility, there will be more than 600 confirmed cases of COVID-19 in Shenzhen.