Reconstruction Of Missing Seismic Data Based On Anti-aliasing POCS Algorithm

In seismic exploration,missing seismic tracks seriously affect processing and interpretation of seismic data,so it is necessary to use effective reconstruction methods to recover missing tracks.For reconstruction of missing seismic tracks with uniform grid in space,most of reconstruction algorithms transform them into solving l₁ minimum norm constraint problem,and Projection Onto Convex Sets is a kind of simple algorithm which can be used to solve the problem effectively.But the traditional algorithm only can solve reconstruction problem of random missing seismic tracks effectively.When regular missing tracks is reconstructed,aliasing is not suppressed in space,affecting resolution of seismic data seriously.For the kind of problem,mask function is introduced into traditional reconstruction process based on FFT and POCS,taking origin of coordinate in f-k spectrum as a starting point,establishing a certain range of angle rays to reconstruct tracks and suppress aliasing.By setting thresholds artificially,identifying the real spectrum ranges,and finally establishing a mask function with the same dimension as data’s spectrum dimension.The value of mask function corresponding to aliasing is zero,real spectrum is one.Based on FFT,anti-aliasing POCS is applied,discussing influences of results corresponding to slope ranges and scan width of rays.Compared with the traditional method,the novel method can reconstruct both random missing data and regular missing data,improving SNR and accelerating processes.However,it is not effective for complex nonlinear events data.Therefore,mask function is introduced into an algorithm based on curvelet transform and POCS,determining boundaries of real spectrums and selecting threshold constantτin detail.Finally,anti-aliasing POCS is applied to the actual data,and result is better,advancing SNR and accelerating convergences.