Ghanavati, M. and Shad Manaman, N., 2022. An efficient automatic Curvelet-Contourlet fault detection method using fuzzy entropy. Journal of Seismic Exploration, 31: 219-238. Accurate detection of faults in seismic data can affect various disciplines such as subsurface geological studies, petroleum exploration, drilling strategies and production. Still, it is in general a manual and time consuming task. Therefore, bringing automatic methods to this field which can accurately extract useful information from seismic images and locate faults would be valuable. In this paper we propose a novel method for automatic seismic fault detection using combination of multiresolution multidirectional algorithms and fuzzy entropy theory. The paper employs Curvelet (CV) and Contourlet (CN) transforms for feature extraction from transformed domain and to capture both detail and smooth information content of the data. The proposed framework introduces a novel feature space by extracting features in temporal domain using CN transform to capture smooth contour information and CV transform to capture details along the curve features in order to improve detection performance. It also introduces an automatic feature selection algorithm using differentiation which highlights fault information, to isolate faults from reflectors adaptively. The reduced coefficients are used as feature vectors to locate faults more accurately. Then, a multi-level thresholding based on fuzzy partition of the histogram and entropy theory is applied to classify image pixels into fault and non-fault. According to results and assessments, this method is very efficient in accurately locating faults and eliminates the need for manually interpret fault surfaces.