With the development and application of GEO Earth observation satellites, the mission planning of GEO imaging satellites has become a new research issue. Due to the operational characteristics and advantages of GEO imaging satellites, GEO satellites are well suited for high frequency revisiting of point targets. This paper mainly studies the working characteristics of GEO imaging satellites and proposes a general GEO satellite imaging working mode. After the appropriate constraint simplification, a constraint satisfaction mathematical model is established. In order to solve the problem better, a genetic algorithm based on heuristic rules (HGA) is designed, through improving its population initialization function, crossover operator and mutation operator. Finally, the rationality and effectiveness of the algorithm are verified by experimental scenarios of different scales.