Different versions of the Radon transform (RT) are widely used in seismic data processing tofocus the recorded seismic events. Multiple separation, data interpolation, and noise attenuationare some of RT applications in seismic processing work-flows. Unfortunately, the conventional RTmethods cannot focus the events perfectly in the RT domain. This problem arises due to theblurring effects of the source wavelet and the nonstationary nature of the seismic data. Sometimes,the distortion results in a big difference between the original data and its inverse transform. Wepropose a nonstationary deconvolutive RT to handle these two issues. Our proposed algorithm takesadvantage of a nonstationary convolution technique. that builds on the concept of block convolutionand the overlap method, where the convolution operation is defined separately for overlapping blocks.Therefore, it allows the Radon basis function to take arbitrary shapes in time and space directions. Inaddition, we introduce a nonstationary wavelet estimation method to determine time-space-varyingwavelets. The wavelets and the Radon panel are estimated simultaneously and in an alternative way.Numerical examples demonstrate that our nonstationary deconvolutive RT method can significantlyimprove the sparsity of Radon panels. Hence, the inverse RT does not suffer from the distortioncaused by the unfocused seismic events.