Introduction

It has long been recognized that the one of the most accurate ways to image areas that are geologically complex is prestack depth migration. Many prestack depth migration algorithms have been developed and applied to common shot gathers or common receiver gathers. This approach, however, is not efficient in computation, especially in 3D. Full prestack depth migration still requires too much time for routine processing even in 2D.

An alternative process to full prestack depth migration is migration of a stack that is based on wavefront reconstruction. Huygens' principle of superposition permits the synthesis of arbitrarily shaped downward propagating wavefronts from many spherical wavefronts. Imaging or processing based on wavefront synthesis has been considered in seismic exploration. Taner (1976) proposed to synthesize plane wave sources at the surface by stacking traces in a common receiver gather. Later, Schultz and Claerbout (1978) showed a way to observe angle-dependent reflection coefficients by synthesizing a slanted plane wave source. Recently, Rietveld et al. (1992) and Rietveld and Berkhout (1992) presented the ``optimum seismic illumination technique'' by synthesizing a plane wave at a specified depth.

This paper reviews the profile imaging technique as a conventional prestack depth migration with its limits and problems. Optimum seismic illumination ( Rietveld, et al, 1992) is reviewed in terms of how it overcomes the hurdles of conventional depth migration and tested with the Marmousi data. Some discussions are given about the problems of the optimum illumination and possible improvements to pursue in the future.