The aim of this work was to compare various dip estimation methods and apply the best of them to identify the dip of seismic interference noise (SI) in the shot domain. Four dip estimation methods were considered: correlation based method, structure tensor estimation, linear and nonlinear plane wave destruction methods. Both synthetic and real data were used for the tests. The methods were compared with respect to their accuracy, sensitivity to noise, runtime and the number of different parameters the user needs to set. The nonlinear plane wave destruction method was chosen as the one which best meets the requirements. This method was then applied to estimate the dip of SI noise in one seismic line. The noise was then succesfully removed by the τ − p - common−p SI attenuation method. Dip based signal and noise separation was also tested, however, it removed too much of the signal.