Seismic Impedance Inversion Based on Sub-Image and Convolutional Neural Network
J. Li, Q. Du, L. Fu and Q. Li
Event name: 81st EAGE Conference and Exhibition 2019
Session: Poster: AI/Digitalization for Interpretation - Various Application
Publication date: 03 June 2019
Info: Extended abstract, PDF ( 535.97Kb )
Price: € 20
Based on image recognition ability of convolutional neural network (CNN), we propose a method of impedance inversion using sub-image element. First, sub-image element of seismic record is extracted as the input while the value of corresponding logging impedance is considered as the output of dataset. Then, based on the training and testing set, impedance profile is obtained by CNN. Finally, the predicted results of the proposed method are compared with those using traditional CNN, BP and Bayesian inversion, which shows that the proposed method can efficiently obtain more accurate and reasonable results only using seismic data.