The K distribution,one of the most important distributions for modeling synthetic aperture radar (SAR) data,shows great capability of fitting different clutter regions,such as the ocean area,the forest and the cropland with growing crops etc.However,the classical method of moments (MoM) for estimating the parameters of the K distribution,which is widely adopted,has the limitations of acquiring the number of looks by prior knowledge of the SAR image,poor estimation veracity and leading the invalidation of the K distribution and so on.Therefore,this paper proposes a fast and robust method of parameter estimation for the K distribution based on the Mellin transform.The novel method has the following advantages:firstly,it gets rid of the problem of invalidating the K distribution;secondly,it regards the number of looks as a parameter to be estimated,like the shape and scale parameters;thirdly,estimation values of the three parameters can be quickly and accurately acquired,all of which guarantee the K distribution’s fitting precision.According to the results of the experiments,the proposed estimator shows better fitting performance than the MoM estimators.