Interpretation of seismic data is a time-consuming and arduous task. Clustering analysis as an intelligent analysis method can be applied to the petroleum industry. While most clustering algorithms have good performance on transactional data, they are not suitable for seismic data. Unlike traditional data, seismic data have some characteristics of its own: spatially position, fuzzy nature and arbitrary shape of clusters. In this paper, we present a clustering algorithm based on fuzzy sets to tackle these problems. Experimental results show that our algorithm has the ability to discover clusters with arbitrary shape and is tolerant to noise. We also show the applicability of the proposed algorithm to real seismic data.