Wavelet phase estimation using ant colony optimization algorithm

Eliminating seismic wavelet is important in seismic high-resolution processing. However, artifacts may arise in seismic interpretation when the wavelet phase is inaccurately estimated. Therefore, we propose a frequency-dependent wavelet phase estimation method based on the ant colony optimization (ACO) algorithm with global optimization capacity. The wavelet phase can be optimized with the ACO algorithm by fitting nearby-well seismic traces with well-log data. Our proposed method can rapidly produce a frequency-dependent wavelet phase and optimize the seismic-to-well tie, particularly for weak signals. Synthetic examples demonstrate the effectiveness of the proposed ACO-based wavelet phase estimation method, even in the presence of a colored noise. Real data example illustrates that seismic deconvolution using an optimum mixed-phase wavelet can provide more information than that using an optimum constant-phase wavelet.