Five-dimensional interpolation in the OVT domain for wide-azimuth seismic data
Liyan Zhang, Ang Li, Jianguo Yang, Shichao Li, Yulai Yao, Fei Xiao, Yiming Huang
Abstract
Abstract To improve the imaging quality of wide-azimuth seismic data and enhance the uniformity of the attributes between adjacent bins, we developed a novel interpolation method in the offset-vector tiles (OVT) domain for wide-azimuth data. The orthogonal matching pursuit (OMP) interpolation method based on the Fourier transform is a frequency-domain processing technique based on discrete Fourier interpolation that achieves the goal of anti-aliasing by extracting the weight factor in the effective band from low-frequency data without aliasing. For data reconstruction, the OMP-based data interpolation technique in the OVT domain comprehensively uses the seismic data in five dimensions: the vertical and horizontal coordinates, time, offset and azimuth. Compared with conventional three-dimensional data interpolation, five-dimensional interpolation in the OVT domain is more accurate and achieves better results in practical applications.