Velocity Analysis By Homeomorphic Image Focusing

In the conventional seismic processing the image of the seismic reflector is obtained by applying a depth migration<br>method to the input seismic data. This is reached by assuming a known macrovelocity model and using a diffraction<br>stack process on the observed seismic data. In this paper a new kind of image of the seismic reflector is defined by<br>considering only the kinematic aspects, within a seismic model consituted by homogeneous multilayers separated by<br>curved interfaces. The referred new image process is performed as solution of the inverse reflection normal ray problem,<br>having as input data a zero-offset seismic section and as searched-for parameters the radius of curvature (RNIP) of the<br>Normal Incidence Point (NIP) wave, the emergence angle (β0 ) and the emergence point x0 of the reflection normal<br>ray . The cited three parameters are then determined for each normal reflection ray by applying the Double Diffraction<br>Stack Technique (DDSI), and the image itself is builted by putting the amplitude of the diffraction stacked data into the<br>center of curvature of the NIP wave, the so-called homeomorphic image. By using three different macrovelocity models,<br>we provide a measure of the sensibility of the proposed imaging process on varying the bottom layer velocity of the<br>target zone.