We propose wavelet-based data reconstruction to interpolate land data with large dynamic range without amplitude processing or windowing. We test two approaches using compressive sampling to recover full unaliased data: sparsity promoting reconstruction by minimization and projection onto convex sets. Unlike the Fourier domain, the wavelet domain provides a good representation of non-stationary signals and allows to rebuild data of high dynamic range with relatively small percentage of all coefficients. We solve an minimization problem to find a sparse representation of full data in the wavelet domain and compare it with results of a wavelet-domain POCS algorithm. Tests on field data reveal that both approaches can recover missing data highly coherent with the existing data, while taking advantage of the full dynamic range of the data.
Presentation Date: Wednesday, September 18, 2019
Session Start Time: 8:30 AM
Presentation Time: 9:20 AM
Location: 221A
Presentation Type: Oral
