Seismic facies recognition based on prestack data using deep convolutional autoencoder

Prestack seismic data carries much useful information that can help us find more complex atypical reservoirs. Therefore, we are increasingly inclined to use prestack seismic data for seismic facies recognition. However, due to the inclusion of excessive redundancy, effective feature extraction from prestack seismic data becomes critical. In this paper, we consider seismic facies recognition based on prestack data as an image clustering problem in computer vision (CV) by thinking of each prestack seismic gather as a picture. We propose a convolutional autoencoder (CAE) network for deep feature learning from prestack seismic data, which is more effective than principal component analysis (PCA) in redundancy removing and valid information extraction. Then, using conventional classification or clustering techniques (e.g. K-means or selforganizing maps) on the extracted features, we can achieve seismic facies recognition. We applied our method to the prestack data from physical model and LZB region. The result shows that our approach is superior to the conventionals.