SeiSIM: Structural Similarity Evaluation for Seismic Data Retrieval

Structural similarity evaluation is a critical step for retrieving existing databases to find matching records for a given seismic data set. The objective is to enable the re-use of historical findings to assist exploration with new survey data. Currently there are very few structural similarity metrics specifically designed for seismic data, especially seismic survey maps. In this paper, we propose a metric that combines texture similarity and geological similarity, which is derived from discontinuity maps. We test the seismic similarity metric in a retrieval application. The successful results indicate that our approach is promising.