Detecting salt domes in seismic images is very important for the exploration of petroleum reservoirs. However, this task is characterized as being time-consuming when performed by human interpreters due to the structural complexity of salt bodies found on different seismic volumes. This work aims at performing a comprehensive evaluation of texture descriptors broadly used in the image processing community when applied to seismic images. A robust multi-scale analysis is conducted in order to assess which features and corresponding parameters are more relevant for salt dome detection according to various patch sizes. Experiments with the Netherlands Offshore F3 Block public dataset  demonstrate that Gabor filter, GLCM and Histogram features-this last, little explored in the seismic image literature-produced the best results. We performed preliminary experiments with a set of seismic images from the aforementioned dataset, using the proposed multi-scale combination of texture features. The results indicate that these descriptors are able to produce satisfying results for clustering algorithms.