Robust salt-dome detection using the ranking of texture-based attributes

The accurate interpretation and analysis of seismic data heavily depends on the robustness of the algorithms used. We focus on the robust detection of salt domes from seismic surveys. We discuss a novel feature-ranking classification model for saltdome detection for seismic images using an optimal set of texture attributes. The proposed algorithm overcomes the limitations of existing texture attribute-based techniques, which heavily depend on the relevance of the attributes to the geological nature of salt domes and the number of attributes used for accurate detection. The algorithm combines the attributes from the Gray-Level Co-occurrence Matrix (GLCM), the Gabor filters, and the eigenstructure of the covariance matrix with feature ranking using the information content. The top-ranked attributes are combined to form the optimal feature set, which ensures that the algorithm works well even in the absence of strong reflectors along the salt-dome boundaries. Contrary to existing salt-dome detection techniques, the proposed algorithm is robust and computationally efficient, and works with small-sized feature sets. I used the Netherlands F3 block to evaluate the performance of the proposed algorithm. The experimental results suggest that the proposed workflow based on information theory can detect salt domes with accuracy superior to existing salt-dome detection techniques.