In this study, a method for retrieving wind speed from synthetic aperture radar (SAR) imagery collected under extreme weather conditions is proposed. The rationale for this approach relies on the fact that, although co-polarized channels exhibit saturation for wind speed > ~20 m/s, the wave growth can be successfully exploited to gather information on wind speed under extreme weather conditions. Hence, in this study, the intrinsic relationship among the wind-wave triplets (wind speed at 10 m above the sea surface, significant wave height (SWH), and peak wave period) is exploited in order to retrieve wind speeds under tropical cyclone conditions. Experiments, undertaken on actual X-band TerraSAR-X (TS-X) SAR images of tropical cyclones (Typhoon Megi, Hurricane Sandy and Hurricane Miram) and using co-located WAVEWATCH-III (WW3) simulations, revealed the robustness of the proposed approach, which resulted in a root mean square error (RMSE) of 2.54 m/s when comparing the retrieved wind speeds with the values from products delivered by the National Oceanic and Atmospheric Administration (NOAA) Hurricane Research Division (HRD). However, the applicability of herein algorithm will be further confirmed at very strong storms.