The airborne transmission path for SARS-CoV-2 is of primary scientific and health-related interest as it could actually involve management, accessibility, use and functionality of many activities, including hospitals), schools, workplaces, factories, transport, sport venues and outdoor environment. It is necessary to develop a sampling and analytical method for virus-laden bioaerosol that could be considered reliable and validated according to ISO/IEC 17025 requirements. The present paper defines sample pretreatments aiming at recover SARS-CoV-2 from glass-fiber and PTFE filters employed by low and high-volume air samplers. Recovery test results focused on the sample concentration step carried out by means of ultracentrifugation are reported as well. Human coronavirus strain OC43 (a surrogate β-coronavirus with the same SARS-CoV-2 particle structure) was used to validate each step of the recovery tests. We found that the elution efficiency of coronavirus OC43 from glass-fiber and quartz filters could be strongly enhanced by using an elution buffer containing up to 40% of fetal calf serum. Moreover, the recovery from PTFE filters is much higher and easier than from glass-fiber filters: for glass-fiber filters a 3 h-shaking phase, followed by a 30 s-vortexing step, are necessary to elute viral infective particles; for PTFE, 60 min-shaking is enough. The effect of suction time on filters could be resumed as follows: sampling durations up to 20 min at a flow rate of 500 L/min do not affect recovery efficiencies from 10 cm glass-fiber filters, whereas the recovery efficiency of infectious virions from 4.7 cm PTFE filters decreases of a factor 2 after 3 h of sampling at a flow rate of 20 L/min. The recovery efficiency of ultracentrifugation turns out to be around 57%. The effect of storage temperature of filters immersed in a transport medium on coronavirus infectivity is assessed as well. Based on the sampling techniques and the analytical methods developed as described in the present study, many field tests were carried out reporting virus concentrations up to 50 genomic copies per cubic meter of air in domestic environment with poor ventilation condition, whereas in hospital wards the detectable concentrations of SARS-CoV-2 were generally lower than 10 genomic copies per cubic meter of air.