In order to further clarify the transmission context of infectious diseases, a SEIRD epidemic transmission network model based on cellular automata, considering the changes in population size and the impact of emergency relief supplies on dynamic contact rate, was proposed to overcome the shortcomings of traditional dynamic transmission models SIR And SEIR in view of the sudden, unexpected and group characteristics of infectious diseases. Firstly, through data analysis, the contact rate function related to the quantity of emergency supplies and population density was constructed, and the dynamic SEIRD infectious disease transmission model was established. Secondly, the unknown parameters in the model were estimated based on the least square estimation optimization algorithm and the iterative fourth-order Runge-Kutta method. Thirdly, LSODA algorithm was used to carry out implicit numerical integration operation to solve the numerical solution of the ordinary differential equations, and cell evolution rules were established through the solved parameter values, and the transmission trend of infectious diseases was displayed in a way based on cellular automata. Finally, the data of COVID-19 in Hubei Province, China from February 1 to April 1, 2020 were used to predict the transmission network of infectious diseases and verify the model. By setting the relevant parameters of the model, the transmission process of infectious diseases was displayed and its transmission trend was visualized, and the sensitivity of the model parameters was analyzed. The prediction results show that the dynamic SEIRD model is effective.