A residual dictionary learning method for footprint removal from seismic data

Acquisition footprint causes serious interference with seismic attribute analysis, which severely hinders accurate reservoir characterization. Therefore, acquisition footprint suppression has become increasingly important in industry and academia. In this work, we assume that the time slice of 3D post-stack migration seismic data mainly comprises two components, i.e., useful signals and acquisition footprint. Useful signals describe the spatial distributions of geological structures with local piecewise smooth morphological features. However, acquisition footprint often behaves as periodic artifacts in the time-slice domain. In particular, the local morphological features of the acquisition footprint in the marine seismic acquisition appear as stripes. As useful signals and acquisition footprint have different morphological features, we can train an adaptive dictionary and divide the atoms of the dictionary into two sub-dictionaries to reconstruct these two components. We propose an adaptive dictionary learning method for acquisition footprint suppression in the time slice of 3D post-stack migration seismic data. To obtain an adaptive dictionary, we use the K-singular value decomposition algorithm to sparsely represent the patches in the time slice of 3D post-stack migration seismic data. Each atom of the trained dictionary represents certain local morphological features of the time slice. According to the difference in the variation level between the horizontal and vertical directions, the atoms of the trained dictionary are divided into two types. One type significantly represents the local morphological features of the acquisition footprint, whereas the other type represents the local morphological features of useful signals. Then, these two components are reconstructed using morphological component analysis based on different types of atoms, respectively. Synthetic and field data examples indicate that the proposed method can effectively suppress the acquisition footprint with fidelity to the original data.

Download Full-text

Retrieving the leaked signals from noise using a fast dictionary learning method

Geophysics ◽

10.1190/geo2021-0243.1 ◽

2021 ◽

pp. 1-86

Author(s):

Wei Chen ◽

Omar M. Saad ◽

Yapo Abolé Serge Innocent Oboué ◽

Liuqing Yang ◽

Yangkang Chen

Keyword(s):

Dictionary Learning ◽

Seismic Data ◽

State Of The Art ◽

Computational Cost ◽

Learning Method ◽

Global Parameter ◽

Radius Parameter ◽

Orthogonalization Method ◽

Value Decomposition ◽

Learned Features

Most traditional seismic denoising algorithms will cause damages to useful signals, which are visible from the removed noise profiles and are known as signal leakage. The local signal-and-noise orthogonalization method is an effective method for retrieving the leaked signals from the removed noise. Retrieving leaked signals while rejecting the noise is compromised by the smoothing radius parameter in the local orthogonalization method. It is not convenient to adjust the smoothing radius because it is a global parameter while the seismic data is highly variable locally. To retrieve the leaked signals adaptively, we propose a new dictionary learning method. Because of the patch-based nature of the dictionary learning method, it can adapt to the local feature of seismic data. We train a dictionary of atoms that represent the features of the useful signals from the initially denoised data. Based on the learned features, we retrieve the weak leaked signals from the noise via a sparse co ding step. Considering the large computational cost when training a dictionary from high-dimensional seismic data, we leverage a fast dictionary up dating algorithm, where the singular value decomposition (SVD) is replaced via the algebraic mean to update the dictionary atom. We test the performance of the proposed method on several synthetic and field data examples, and compare it with that from the state-of-the-art local orthogonalization method.

Download Full-text

A dictionary learning method for seismic data compression

Geophysics ◽

10.1190/geo2020-0948.1 ◽

2021 ◽

pp. 1-83

Author(s):

Mohammed Outhmane Faouzi Zizi ◽

Pierre Turquais

Keyword(s):

Data Storage ◽

Compression Ratio ◽

Dictionary Learning ◽

Seismic Data ◽

Superior Performance ◽

Seismic Survey ◽

Learning Method ◽

Data Set ◽

Wide Range ◽

Marine Seismic

For a marine seismic survey, the recorded and processed data size can reach several terabytes. Storing seismic data sets is costly and transferring them between storage devices can be challenging. Dictionary learning has been shown to provide representations with a high level of sparsity. This method stores the shape of the redundant events once, and represents each occurrence of these events with a single sparse coefficient. Therefore, an efficient dictionary learning based compression workflow, which is specifically designed for seismic data, is developed here. This compression method differs from conventional compression methods in three respects: 1) the transform domain is not predefined but data-driven; 2) the redundancy in seismic data is fully exploited by learning small-sized dictionaries from local windows of the seismic shot gathers; 3) two modes are proposed depending on the geophysical application. Based on a test seismic data set, we demonstrate superior performance of the proposed workflow in terms of compression ratio for a wide range of signal-to-residual ratios, compared to standard seismic data methods, such as the zfp software or algorithms from the Seismic Unix package. Using a more realistic data set of marine seismic acquisition, we evaluate the capability of the proposed workflow to preserve the seismic signal for different applications. For applications such as near-real time transmission and long-term data storage, we observe insignificant signal leakage on a 2D line stack when the dictionary learning method reaches a compression ratio of 24.85. For other applications such as visual QC of shot gathers, our method preserves the visual aspect of the data even when a compression ratio of 95 is reached.

Download Full-text

A Dictionary Learning Method for Seismic Data Compression

10.3997/2214-4609.202034036 ◽

2020 ◽

Author(s):

M.O. Faouzi Zizi ◽

P. Turquais

Keyword(s):

Data Compression ◽

Dictionary Learning ◽

Seismic Data ◽

Learning Method ◽

Seismic Data Compression

Download Full-text

A multi-task transfer learning method with dictionary learning

Knowledge-Based Systems ◽

10.1016/j.knosys.2019.105233 ◽

2020 ◽

Vol 191 ◽

pp. 105233 ◽

Cited By ~ 2

Author(s):

Xin Zheng ◽

Luyue Lin ◽

Bo Liu ◽

Yanshan Xiao ◽

Xiaoming Xiong

Keyword(s):

Transfer Learning ◽

Dictionary Learning ◽

Learning Method ◽

Task Transfer

Download Full-text

Non-parallel dictionary learning for voice conversion using non-negative Tucker decomposition

EURASIP Journal on Audio Speech and Music Processing ◽

10.1186/s13636-019-0160-1 ◽

2019 ◽

Vol 2019 (1) ◽

Author(s):

Yuki Takashima ◽

Toru Nakashika ◽

Tetsuya Takiguchi ◽

Yasuo Ariki

Keyword(s):

Dictionary Learning ◽

Computational Cost ◽

Tensor Decomposition ◽

Gaussian Mixture ◽

Voice Conversion ◽

Specific Information ◽

Learning Method ◽

Tucker Decomposition ◽

Parallel Data ◽

High Computational Cost

Abstract Voice conversion (VC) is a technique of exclusively converting speaker-specific information in the source speech while preserving the associated phonemic information. Non-negative matrix factorization (NMF)-based VC has been widely researched because of the natural-sounding voice it achieves when compared with conventional Gaussian mixture model-based VC. In conventional NMF-VC, models are trained using parallel data which results in the speech data requiring elaborate pre-processing to generate parallel data. NMF-VC also tends to be an extensive model as this method has several parallel exemplars for the dictionary matrix, leading to a high computational cost. In this study, an innovative parallel dictionary-learning method using non-negative Tucker decomposition (NTD) is proposed. The proposed method uses tensor decomposition and decomposes an input observation into a set of mode matrices and one core tensor. The proposed NTD-based dictionary-learning method estimates the dictionary matrix for NMF-VC without using parallel data. The experimental results show that the proposed method outperforms other methods in both parallel and non-parallel settings.

Download Full-text

Simultaneous inversion of Q and reflectivity using dictionary learning

Geophysics ◽

10.1190/geo2020-0095.1 ◽

2021 ◽

pp. 1-69

Author(s):

Jie Shao ◽

Yibo Wang

Keyword(s):

Conventional Method ◽

Dictionary Learning ◽

Computational Efficiency ◽

Synthetic Data ◽

Inversion Method ◽

Learning Method ◽

Q Value ◽

Value Set ◽

Simultaneous Inversion ◽

Convolution Model

Quality factor ( Q) and reflectivity are two important subsurface properties in seismic data processing and interpretation. They can be calculated simultaneously from a seismic trace corresponding to an anelastic layered model by a simultaneous inversion method based on the nonstationary convolution model. However, the conventional simultaneous inversion method calculates the optimum Q and reflectivity based on the minimum of the reflectivity sparsity by sweeping each Q value within a predefined range. As a result, the accuracy and computational efficiency of the conventional method depend heavily on the predefined Q value set. To improve the performance of the conventional simultaneous inversion method, we have developed a dictionary learning-based simultaneous inversion of Q and reflectivity. The parametric dictionary learning method is used to update the initial predefined Q value set automatically. The optimum Q and reflectivity are calculated from the updated Q value set based on minimizing not only the sparsity of the reflectivity but also the data residual. Synthetic data and two field data sets were used to test the effectiveness of our method. The results demonstrated that our method can effectively improve the accuracy of these two parameters compared to the conventional simultaneous inversion method. In addition, the dictionary learning method can improve computational efficiency up to approximately seven times when compared to the conventional method with a large predefined dictionary.

Download Full-text