High Frequency Imaging (HFI®)
High Frequency Imaging (HFI®), unlike the conventional convolutional model, transforms the seismic trace into a frequency vector space (Nw) and then rotates the seismic vector to maximize the high frequency component in the data. HFI® assumes the noise in the data is Gaussian and is approximately spherical around the origin in the transformed vector space. As the trace vector rotates toward higher frequencies the energy is redistributed to produce a broader spectrum. The HFI® rotation leaves the noise sphere unchanged. Therefore, the signal emerges from beneath the noise level and the high frequencies recovered are usable signal.
The seismic images below in time space compare the input data with the results of HFI® and conventional Spectral Balance or Spectral Whitening. It clearly shows HFI® recovers higher frequencies from low frequencies and thus provides higher resolution events without increasing the noise level. On the contrary, Spectral Balance broadens the bandwidth but not as effective as HFI® and in addition, it raises the noise level and thus makes interpretation more difficult.
Applications
HFI® can be applied before or after stack. It not only improves the seismic resolution for structural, stratigraphic and facies interpretation but also enhances prestack analysis and reservoir evaluation such as:
- HDHR Velocity Analysis for more accurate and higher resolution velocity field
- Conventional and advanced AVO/AVA attributes for derisking tuning problem
- Seal Capacity Cube and Pore Pressure prediction for seal integrity, well planning and drilling hazard prevention
- RockRes® three-term simultaneous prestack inversion for lithology and fluid prediction
 Geotrace and Reliance Work on Improving Seismic Resolution (Oil & Gas Eurasia, January 29, 2008)
High Frequency Imaging in an Exploitation Production Environment - The Murzuq Basin, Case History
 HFI®
|
