Automatic interpretive image-focusing analysis

Abstract

The focusing of a seismic image is directly linked to the accuracy of the velocity model. Therefore, a critical step in a seismic imaging workflow is to perform a focusing analysis on a seismic image to determine velocity errors. Although the offset/aperture-angle axis is frequently used for focusing analysis, the physical axes of seismic images tend to be ignored because the focusing analysis of geologic structures is highly interpretive and difficult to automate. We have developed an automatic data-driven approach using convolutional neural networks to automate image-focusing analysis. Using focused and unfocused geologic faults, our approach makes use of spatial and offset/angle focusing information to robustly estimate local focusing errors within seismic images. The application of our method to a 2D limited-aperture image from the Gulf of Mexico shows that it can correctly estimate local focusing errors within the image, while a traditional semblance-based approach fails. Additionally, the proposed approach has the added benefit of improving the interpretation of faults within the image.