Conventional ground-based ISAR is restrained by the noncooperative motion of targets and has long coherent processing interval (CPI), so its imaging performance is unstable. In this work airborne ISAR imaging technique will be discussed, which can shorten the distance from radar to air target, decrease the CPI, and improve imaging applicability. Firstly, the relative motion between radar and the air target is analyzed. Secondly, the imaging model is built and a whole matched filtering (WMF) imaging algorithm is proposed. Finally, the multi-focus processing is expatiated, the radar flight error is analyzed and the method that adjusting radar flight perimeters to improve the reality imaging quality is discussed.