Regular Multi-Shot Subsampling and Reconstruction on 3D Orthogonal Symmetric Seismic Grids via Compressive Sensing

• Published in: Symposium of Image, Signal Processing, and Artificial Vision pp. 1 - 5
• Main Authors: Lopez, Kareth Leon, Ramirez, Juan M., Agudelo, William, Arguello Fuentes, Henry
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2019
• Subjects: compressive sensing; Geometry; Image reconstruction; Receivers; Seismic acquisition; Seismic measurements; Signal to noise ratio; subsampling; symmetric grids; Three-dimensional displays; Two dimensional displays
• ISSN: 2329-6259
• DOI: 10.1109/STSIVA.2019.8730279

Compressive sensing (CS) principles have been successfully tested to seismic exploration, allowing the development of a new acquisition framework. CS-based seismic acquisition approaches attempt to exploit CS principles to reduce turnaround time and budget of the exploration. Different works have focused on receiver-subsampling, nevertheless, sources in a seismic survey are more expensive than receivers. This work extends the concept of subsampling to an acquisition scheme that consists of multiple missing sources on a 3D orthogonal seismic survey. The proposed acquisition scheme removes multiple sources at regular intervals providing flexibility to the acquisition design. The measurement matrix is modeled and the inverse problem to recover the seismic data from the available measurements is formulated. Numerical experiments on real and synthetic datasets show the promising of CS for multiple source subsampling, resulting in accurate reconstructions with SNR values of around 24 dB.