Objective The sampling method of virus on food packaging surfaces was optimized. The virus recovery rate on different material surfaces was studied by simulating different scenes of virus contamination on the surfaces of frozen food packaging materials.Methods The virus was evenly coated on the surfaces of different materials in different scenarios. Different buffers and different sampling methods were used for sampling residual virus. The virus recovery rate was calculated by virus titer （TCID₅₀）.Results Among the three sample buffers， Hank’s balanced salt solution had the highest virus recovery rate， followed by PBS buffer and MEM media. Among the five sampling methods， the highest recovery was to apply the wet swab once on the surface and then rotated for the next swabing. Using the optimized sampling method to sample virus from the surfaces of different materials， the recovery rates of smooth plastic surface and metal surface were 10.47% and 3.55%， while the living virus of hard plastic surface， thick cardboard surface， kraft paper and wood surface could not be collected. The virus recovery rates of contaminated food on smooth plastic surface before and after frozen were 92.03% and 4.83%， and that of and metal surface were 1.27% and 4.94%. In addition， the recovery rate of hard plastic surface was 2.59% in contaminated food before frozen， and the recovery rate of thick cardboard surface was 1.27% only in the contaminated food after frozen.Conclusion The virvs recovery rates varies depending on the substrate of surfaces. This sampling method is suitable for collecting virus from smooth "hard surfaces". The residual virus on frozen food packaging surfaces can be effectively collected by the sampling method optimized above as well.