Generalized Matched Filtering for Improving Seismic Section Clarity: Incorporating Frequency‐Based Regularization

ABSTRACT Seismic data resolution is crucial for subsurface imaging. But factors such as the finite bandwidth of seismic sources, wavelet interference and noise often degrade the clarity of reflectors, making their detection challenging. Although recent advances in seismic data processing methods have shown promise in enhancing resolution, they remain limited by underlying physical constraints and inherent assumptions. In contrast, matched filtering, a physics‐based signal‐processing technique, can enhance the detectability of known waveforms by maximizing the signal‐to‐noise ratio. In this article, the source wavelet is first approximated using Wiener inverse filtering, followed by blue reflectivity modifications consisting of amplitude compensation of 3 dB/octave and a + π /4 phase correction. After applying the proposed generalized matched filter (GMF), an additional + π /2 phase correction is introduced as a direct consequence of the derivation, ensuring proper phase alignment. Unlike conventional matched filtering, which is designed for isolated reflectors, the GMF developed in this study explicitly incorporates the thin‐layer response by modifying the original matched‐filter equation. This modification represents a key novelty of the study, as it enables the filter to account for the differential effects of the seismic wavelet. To stabilize the results and suppress noise amplification, frequency‐based regularization differentiation is applied. The proposed GMF workflow is validated using stationary and non‐stationary synthetic examples as well as a field seismic dataset. The results demonstrate improved thin‐bed resolution and reflector coherence through wavelet compression and phase correction within the available frequency bandwidth, rather than through artificial spectral broadening. Quantitative analyses based on side‐lobe energy ratio (SLER), power spectral density (PSD), tuning‐thickness measurements and seismic coherence demonstrate that GMF enhances seismic interpretation while preserving the spectral characteristics of the original data.