Routine seismic imaging algorithms often require velocity models. Therefore, accuracy of velocity models obtained by velocity analysis plays a crucial role in seismic imaging. Seismic velocity analysis in CMP domain has been dominated by the use of a powerful coherency measurement tool, called semblance. Although it works quite reasonable for the majority of practical cases, it is incapable of dealing with polarity reversals across moveout curves caused by class II AVO effects. In this research, we proposed an inversion-based velocity analysis algorithm that is based on fuzzy c-means clustering, which has been recently considered in geophysical applications. We apply the proposed algorithm to a synthetic seismic CMP gather and compare the corresponding outcomes with the results obtained by classic semblance analysis. The results suggest the effectiveness of the proposed algorithm in the case of polarity reversals along moveout curves.

Open-Access Online Publication: March 03, 2023