The goal of this paper is to apply the algorithm of Felzenszwalb and Huttenlocher (2004), introduced above, to seismic data. Publicly-available code from Felzenszwalb (2010) allows for relatively easy implementation for standard images; however, seismic images are very different from photographs or other types of images. The following sections describe the rationale and procedure for modifying and adding additional features to the algorithm in order to apply it to seismic data.