Fisher Discriminant Analysis (FDA), a supervised feature reduction method in seismic object detection

Abstract

Automatic processes on seismic data using pattern recognition is one of the interesting fields in geophysical data interpretation. One part is the seismic object detection using different supervised classification methods that finally has an output as a probability cube. Object detection process starts with generating a pickset of two classes labeled as object and non-object and then selecting a set of attributes that are inputs to a classifier. As a crucial step before classification, a feature extraction algorithm shall be implemented to transfer data from input space to feature space resulting in dimensionality reduction. In this paper, two feature extraction methods Principal Component Analysis (PCA) and Fisher Discriminant Analysis (FDA) in seismic object detection are compared. It is aimed to study fluid migration pathways in the North Sea and SVM classifier is used for classification thereafter. Finally, the obtained results show that in FDA classification error is less than PCA. The second and most important result is posterior probability in physical domain that in FDA is better and more interpretable than PCA.