An image-guided method for automatically picking common-image-point gathers

Using common-image-point gathers (CIPs) instead of conventional image gathers for migration velocity analysis is an efficient approach for analyzing migrated images at locations that are related to geological structures. This approach reduces computational costs and may improve migration velocity analysis by acounting for diving energy. However, manually picking CIP locations is tedious and time-consuming, especially for 3D images; therefore, an automated method for picking CIPs is desirable. To take advantage of the potential computational cost savings, CIPs should be constructed at relatively sparse locations throughout an image. Furthermore, to facilitate improved analysis, CIPs should be constructed along geological features. We provide a new method for automatically picking CIP locations from seismic images. This method uses local image properties including planarity, structure-oriented semblance, and the amplitude envelope to compute a priority-map (PMAP) of seismic images, where higher PMAP values indicate better CIP locations. After the priority map is computed, CIP locations are picked sparsely throughout the image using a greedy heuristic.