Abstarct: One of the most important geological structures in exploratory studies are faults. Faults create oil traps by displacing the stratification. Therefore, the study of fault structures is important. Sometimes due to structural complexity and noise in seismic data, identifying the fault plain is a difficult procedure and it is not possible to determine their exact location from seismic data. Seismic attributes are tools that reveal the information contained in seismic data and allow the interpreter to interpret qualitatively and quantitatively. However, the presence of noise and the dependence of seismic attributes on some parameters such as slope and azimuth of structures, makes it sometimes impossible to detect the fault pattern using only one or more attributes. The other method to detect fault pattern is using edge detection filters which are efficient in identifying the edge of discontinuities. In this thesis, by using a combination of seismic attributes and edge detection methods faulty structures have been identified on seismic data. Due to the fact that the basis of edge detection methods is derivation, they also increase image noise along with highlighting fault structures. Therefore, to prevent this problem, mathematical morphology operations are performed before applying the edge detection operation on the seismic attribute cubes to reduce image noise and highlight faulty structures. In the next step, the edge detection operators are applied to the attribute cubes. In the final step, the results obtained from each stage of the process are combined using color blending (RGB) and interpreted. The results of the study show the high ability of the proposed strategy in identifying fault structures.