Coherence index and curvelet transformation for denoising geophysical data

Area of Science:

• Geophysics
• Signal Processing
• Data Analysis

Background:

• Geophysical data, particularly seismic data, often contain stochastic noise, such as ground roll (GR), which obscures valuable geologic information.
• The removal of coherent noise is a fundamental challenge in processing geophysical datasets, impacting the interpretation of subsurface structures.
• Existing methods for noise reduction may be computationally intensive or lead to significant distortion of useful data.

Purpose of the Study:

• To develop and present an efficient and effective denoising method for geophysical data contaminated with coherent noise.
• To apply the proposed method to seismic data corrupted by ground roll (GR) noise.
• To demonstrate the method's ability to preserve essential geologic information while removing noise.

Main Methods:

• A novel denoising approach is proposed, leveraging the curvelet transformation (CT) for its optimal representation of edge discontinuities.
• The method involves computing a coherence index (CI) to identify noisy regions within the data.
• The curvelet transformation (CT) is then applied to denoise the identified contaminated segments.

Main Results:

• The proposed method successfully removes ground roll (GR) noise from seismic data with minimal distortion to the non-contaminated areas.
• The algorithm demonstrates significant computational efficiency compared to previous denoising techniques.
• Successful application to both synthetic and real seismic data for hydrocarbon reservoir analysis is shown.

Conclusions:

• The curvelet transformation (CT)-based denoising method offers an effective solution for removing coherent noise like ground roll (GR) from geophysical data.
• This approach provides a computationally efficient alternative for seismic data processing, enhancing the extraction of geologic information.
• The method holds promise for improving the accuracy and reliability of hydrocarbon reservoir exploration.