Adaptive windowing algorithms are critical for improving imaging speed with the Gabor wavefield extrapolator. These algorithms help to reduce redundancy of computation while keeping a certain level of imaging accuracy. We propose and test a new adaptive windowing algorithm for the Gabor depth imaging. The new algorithm uses a phase error criterion in the Gabor extrapolation algorithm to determine a set of spatially variable windows that sum to unity. The new windowing algorithm incorporates an important physical constraint because the phase is one of the basic quantities used to describe propagating wavefields. The phase error windowing algorithm is indirectly dependent upon both velocity and wavenumber since the phase is a function of both variables. Comparisons between the new algorithm and a previously published method using velocity alone as a criterion are presented.
View full article as PDF (0.60 Mb)