Full text loading...
-
Co-core Trace Transform Based Radial Trace Time-frequency Peak Filtering for Seismic Random Noise Attenuation
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 76th EAGE Conference and Exhibition 2014, Jun 2014, Volume 2014, p.1 - 5
Abstract
The time frequency peak filtering (TFPF) algorithm is an effective tool in random noise attenuation for its advantages in non-stationary signal estimation, especially when the prior knowledge is unknown. Testing of TFPF has made it possible to recover the seismic reflected signals in noise level down to even -9dB. The radial trace TFPF (RT-TFPF) modifies the filtering with the help of the radial trace transform and improves the performance both in noise attenuation and seismic signal preservation. However, the RT-TFPF allows the unbiased estimation only when the radial traces approximately align with the reflection event directions. Conventional RT-TFPF with fixed direction can not satisfy this condition everywhere. Therefore, a Co-Core trace transform based RT-TFPF with temporal-spatial directions is proposed in this paper. The proposed algorithm utilizes the similarity of data on the reflection event and computes the temporal-spatial radial directions by seeking for the local maximum value point of the Co-Core trace function. With the proposed method, strong random noise is attenuated, while at the same time, the estimation of the reflected signal is more accurate, making the recovery of reflection events more continuous and smooth.