One major difficulty in land seismic data processing is noise related to a complex near surface. The presence of sand dunes and loose deposits overlaying relatively high velocity layers leads to the generation of strong near-surface noise that travels at low velocities. This energy is often aliased despite the small shot and geophone spacing of modern 3D wide-azimuth seismic surveys. Direct application of velocity-based filtering in frequency-wavenumber (f-k) or Radon domains is not effective in suppressing this noise. Seismic data need to be interpolated and regularized to dealias the slow-traveling near-surface arrivals before denoising technologies are applied. In this work, we utilized an f-k domain seismic data interpolation technique similar to the antileakage Fourier transform method for its ability to remove leaked energy in the f-k domain arising from gaps and irregularities during data acquisition. To deal with aliasing in the input data, we implemented an antialiasing weighting function based on piecewise linearity of the seismic data. Seismic images obtained from processing with data regularization show significant signal-to-noise ratio improvement over previous results. Because of the uplift brought by this new interpolation technology and workflow, data interpolation and regularization have become an integral part of Saudi Aramco's land 3D data processing workflow.