Seismic Data Compression Research Articles

Deep learning model for fiber optic distributed acoustic sensing seismic data compression based on autoencoder and recurrent neural networks

Because of the extremely high sampling rate of fiber optic distributed acoustic sensing (DAS) equipment, the amount of DAS seismic data collected is enormous, which poses great challenges to the transmission and storage of DAS seismic data. Therefore, it is essential to study the compression methods of DAS seismic data. Existing data compression methods such as wavelet transform, cosine transform, and convolutional autoencoder (CAE) still have room for improvement in the compression performance and compression ratio (CR). Thus, we have proposed what we believe to be is a novel deep learning model called the recurrent autoencoder (RAE) for high-performance compression of DAS seismic data. Under different CRs, we have designed performance evaluation experiments for RAE models based on different RNN modules with different loss functions. When the CR of the RAE model is 8, the signal-to-noise ratio (SNR) of the reconstructed DAS seismic data reaches 40.60 dB, which is better than that of the CAE’s 11.58 dB. The ultimate CR was increased to 512 without reducing the compression quality, which is 4.12 times higher than the CAE model. The LSTM with a weighted loss function improves the SNR to 43.66 dB at a CR of 8, which is 3.06 dB higher than the LSTM with additive loss function. The results show that the RAE model proposed with a weighted loss function in this paper has excellent DAS seismic data compression performance and provides a high CR, which can be widely applied in large-scale DAS seismic data compression.

Composite loss function for 3-D poststack seismic data compression

This work introduces composite functions to compute distortion in volumetric seismic data. Several loss functions, such as those based on Lp-functions, ignore the structure of 3-D seismic data, treating it as a unidimensional vector. Alternatively, applying distinct functions in each axis, properly designed for dimension reduction, can evaluate seismic data error according to its unity and magnitude. We thus propose a novel multidimensional composite loss function to evaluate through dimension reduction, suitable for seismic data compression within a method named 3DSC-GAN. It replaces the usual peak signal-to-noise ratio (PSNR) metric as the distortion function. An extensive study is conducted to analyze potential combinations of functions for the 3-D poststack seismic data compression problem. Results indicate that the new function contributes to improve the neural network learning step. Our method provides superior reconstructions, both quantitatively and qualitatively, when compared to the PSNR metric.

Combining PCA and DCT for Improved Passive Seismic Data Compression

Combining PCA and DCT for Improved Passive Seismic Data Compression

Dual‐sensor wavefield separation in a compressed domain using parabolic dictionary learning

AbstractIn the marine seismic industry, the size of the recorded and processed seismic data is continuously increasing and tends to become very large. Hence, applying compression algorithms specifically designed for seismic data at an early stage of the seismic processing sequence helps to save cost on storage and data transfer. Dictionary learning methods have been shown to provide state‐of‐the‐art results for seismic data compression. These methods capture similar events from the seismic data and store them in a dictionary of atoms that can be used to represent the data in a sparse manner. However, as with conventional compression algorithms, these methods still require the data to be decompressed before a processing or imaging step is carried out. Parabolic dictionary learning is a dictionary learning method where the learned atoms follow a parabolic travel time move out and are characterized by kinematic parameters such as the slope and the curvature. In this paper, we present a novel method where such kinematic parameters are used to allow the dual‐sensor (or two‐components) wavefield separation processing step directly in the dictionary learning compressed domain for 2D seismic data. Based on a synthetic seismic data set, we demonstrate that our method achieves similar results as an industry‐standard FK‐based method for wavefield separation, with the advantage of being robust to spatial aliasing without the need for data preconditioning such as interpolation and reaching a compression rate around 13. Using a field data set of marine seismic acquisition, we observe insignificant differences on a 2D stacked seismic section between the two methods, whereas reaching a compression ratio higher than 15 when our method is used. Such a method could allow full bandwidth data transfer from vessels to onshore processing centres, where the compressed data could be used to reconstruct not only the recorded data sets, but also the up‐ and down‐going parts of the wavefield.

1-ADM-CNN: A Lightweight In-field Compression Method for Seismic Data

Large-scale seismic acquisition, versatility, flexibility, automation, and scalability are the objectives of future oil and gas exploration technology. An example of emerging technology for seismic monitoring is distributed acoustic sensing (DAS). The significant amount of data produced by DAS is a challenge that necessitates the development of new technologies for its efficient handling and processing. A typical seismic survey, on the one hand, can generate hundreds of terabytes of raw seismic data per day. The demand for wireless seismic data transmission, on the other hand, remains enormous. The massive amount of data transmission from geophones to the on-site data collection center and its storage poses significant challenges. A lightweight compression procedure is required in order to reduce the data traffic and the storage size at the data center without putting an extra burden on a geophone. In this brief, an efficient implementation of a 1D convolutional neural network (CNN) together with 1-bit adaptive delta modulation is presented for in-field seismic data compression. It is worth mentioning here that the training of CNN is done offline on synthetic data set and hence, the proposed approach has potential for real-time implementation. Furthermore, no assumption on the underlying statistics of noise or the seismic signal is imposed and consequently, the proposed method is suitable for a wide range of seismic data. Furthermore, the proposed method works in the time-domain, unlike existing transform-domain methods, making it suitable for quick diagnosis of bad traces at the data center. Simulation results with real data set reveal that the proposed approach achieves a signal-to-noise ratio (SNR) of approximately 30 dB with a compression gain of 35: 1. Finally, significant superiority in terms of compression gain and reconstruction quality is demonstrated when compared to the existing methods.

A dictionary learning method for seismic data compression

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 (DL) 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 DL-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, and (3) two modes (i.e., two different parameterizations) are proposed depending on the geophysical application. Based on a test seismic data set, we have determined superior performance of our workflow in terms of compression ratio (CR) for a wide range of signal-to-residual ratios, 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 our 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 DL method reaches a CR higher than 20. For other applications, such as visual quality control of shot gathers, our method preserves the visual aspect of the data even when a CR of 95 is reached.

Poststack Seismic Data Compression Using a Generative Adversarial Network

This work presents a method for volumetric seismic data compression by coupling a 3-D convolution-based autoencoder to a generative adversarial network (GAN). The main challenge of 3-D convolutional autoencoders for data compression is how to fully exploit volumetric redundancy while keeping reasonable latent representation dimensions. Our method is based on a convolutional neural network for seismic data compression called 3DSC. Its encoder and decoder use 3-D convolutions and are connected by a latent representation with the same dimensions as its 2-D network counterparts. Our main hypothesis is that the 3DSC architecture can be improved by adversarial training. We, thus, propose a new 3-D-based seismic data compression method (3DSC-GAN) by coupling the 3DSC network to a GAN. The seismic data decoder is used as a generator of poststack data that are integrated with a discriminator module to better exploit 3-D redundancy. Results show that our method outperforms previous seismic data compression methods for very low target bit rates, increasing the peak signal-to-noise ratio (PSNR) with fairly high visual quality.

3-D Poststack Seismic Data Compression With a Deep Autoencoder

We approach the problem of 3-D poststack seismic data compression by training a model based on a deep autoencoder. Our network architecture is trained to consider the similarity between 3-D seismic sections drawn from one or multiple seismic volumes. A whole seismic volume is compressed with the latent representations of each of its composing volumetric sections. The goal is to compress the seismic data at very low bit rates with high-quality reconstruction. Our model is suitable for training general compressors from multiple seismic surveys or for specialized compression of a single seismic volume. Results show that our method can compress seismic data with extremely low bit rates, below 0.3 bits-per-voxel (bpv) while yielding peak signal-to-noise ratio (PSNR) values over 40 dB.

Seismic Data Compression Using Deep Learning

The exponential growth of the size of seismic data recorded in seismic surveys and real time data monitoring makes seismic data compression strongly demanded. Furthermore, compression will lead to an efficient use of the bandwidth assigned for the communication link between the seismic stations and the main center. In this paper, two convolutional autoencoders (CAEs) are proposed for seismic data compression. The two algorithms are mainly based on the convolutional neural network (CNN), which has the capability to compress the seismic data into feature representations, thereby allowing the decoder to perfectly reconstruct the input seismic data. The results show that the first model is efficient at low compression ratios (CRs), while the second model improves the signal-to-noise ratio (SNR) from about 3 dB to 12 dB compared to the other benchmark algorithms at moderate and high CRs.

Hardware-software implementation of the compression and transmission of seismic data

A real-time wireless network architecture for seismic data ac-quisition system based on multi-level radio network is proposed. The sin-gle-chip programmable transceiver Si446x that allows creating the first level of the network is considered. Physical layer with GFSK modulation in the sub-gigahertz band is considered. Data link layer implementation with generic frame structure for a network, operating in synchronous mode, is presented. An algorithm for seismic data accumulating and com-pressing is presented. The comparison of existing and proposed solutions is given.

High Bit-Depth Seismic Data Compression: A Novel Codec Under the Framework of HEVC

Motivated by the superior performance of High Efficiency Video Coding (HEVC), and driven by the rapid growth in data volume produced by seismic surveys, in this work we explore a 32 bits per pixel (b/p) extension of the HEVC codec for compression of seismic data. We propose to reassemble seismic slices in a format that corresponds to video signal and benefit from the coding gain achieved by HEVC inter mode, besides the possible advantages of the (still image) HEVC intra mode. To this end, we modify almost all components of the original HEVC codec to cater for high bit-depth coding of seismic data: Lagrange multiplier used in optimization of the coding parameters has been adapted to the new data statistics, core transform and quantization have been reimplemented to handle the increased bit-depth range, and modified adaptive binary arithmetic coder has been employed for efficient entropy coding. Even though the new codec after implementation of the proposed modifications goes beyond the standardized HEVC, it still maintains a generic HEVC structure, and it is developed under the general HEVC framework. Thus, we tailored a specific codec design which, when compared to the JPEG-XR and commercial wavelet-based codec, significantly improves the peak-signal-to-noise-ratio (PSNR) vs. compression ratio performance for 32 b/p seismic data. Depending on a configuration, PSNR gain goes from 3.39 dB up to 9.48 dB. Also, relying on the specific characteristics of seismic data, we proposed an optimized encoder that reduces encoding time by 67.17% for All-I configuration on trace image dataset, and 67.39% for All-I, 97.96% for P2-configuration and 98.64% for B-configuration on 3D wavefield dataset, with negligible coding performance losses.

A fidelity-restricted distributed principal component analysis compression algorithm for non-cable seismographs

This paper proposes a distributed principal component analysis seismic data compression algorithm based on fidelity restriction for non-cable seismic exploration systems. The proposed method compresses data effectively according to a customized fidelity. It first determines the cumulative contribution rate, then calculates the global eigenvector matrix, generates a projection in the seismographs and finally transmits the projected data. The numerical simulation results show that when the total waveform fidelity reaches 92%, the compression ratio still stands at 77 times or more. Furthermore, the field experiment also verifies the performance of the algorithm.

Deep Neural Networks with Extreme Learning Machine for Seismic Data Compression

Advances on seismic survey techniques require a large number of geophones. This leads to an exponential growth in the size of data and prohibitive demands on storage and network communication resources. Therefore, it is desirable to compress the seismic data to the minimum possible, without losing important information. In this paper, a stacked auto-encoder extreme learning machine (AE-ELM) for seismic data compression is proposed. First, a deep asymmetric auto-encoder is constructed, in which nonlinear activation functions are used in the encoder hidden layers and linear activation functions are utilized in the decoder layers. Second, the encoder hidden layers are connected in a cascade way, so that outputs of a hidden layer are considered as the inputs to the succeeding hidden layer. Third, the optimal weights of connections between the layers of the decoder are solved analytically. Lastly, the AE-ELMs are stacked to create the complete encoder/decoder. The extreme learning machine (ELM) is selected due to its analytical calculation of weights efficient training that is suitable for practical implementation. In this neural network, data compression is achieved by transforming the original data through the encoder layers where the size of outputs from the last encoder hidden layer is smaller than the original data size. The proposed method exhibits a comparable reconstruction quality on a real dataset but with a much shorter training duration than other deep neural networks methods. This neural network with more than 8000 hidden units achieved $$ 1.28 \times 10^{ - 3} $$ of normalized mean-squared error for 10:1 of compression ratio with only 8.23 s of training time.

Multiscale Sparse Dictionary Learning With Rate Constraint for Seismic Data Compression

Seismic survey is one of the most effective tools for oil and gas exploration. To date, there has been an exponential growth in the size of seismic data required for large-scale seismic survey. For transmission and storage purposes, we propose a novel seismic compression method. First, a multiscale sparse dictionary learning model with rate constraint is presented. By combining the advantages of multiscale decomposition and dictionary learning, the seismic data could be effectively represented as a sparse matrix. Rate constraints are used to obtain the sparse coefficients that are properly tailored to the compression objective. To solve the optimization problem, the alternating direction method of multipliers is adopted. Furthermore, a seismic compression scheme based on the learned dictionary is introduced. Finally, public seismic datasets are used to verify the efficiency of different seismic data compression methods. The experimental results indicate that the proposed method achieves the best seismic compression performance, including rate-distortion tradeoff and visual quality.

Atom‐based‐segmented MP compression algorithm for seismic data

Data compression is an effective way to improve the seismic data transmission efficiency. The features of seismic exploration are long sampling time and large data quantity, so the compression algorithm should achieve high-fidelity, high-compression ratio (CR) and low-compression time. Since the existing compression algorithms cannot meet the requirements of site-collected seismic data compression, the segmented matching pursuit (SMP) compression algorithm based on new atom dictionary is proposed. This method is based on the principle of MP. A novel-segmented compression structure is adopted. The modified Morlet wave atom dictionary is designed to replace the previous dictionaries. The results of comparative experiments show that the proposed algorithm makes improvements in CR and compression time with the same fidelity requirement.

Advances in Seismic Data Compression via Learning from Data: Compression for Seismic Data Acquisition

The next generation of oil and gas exploration technology is moving toward large-scale seismic acquisition, automation, and flexibility. This phenomenon has accelerated the interest in moving away from traditional seismic acquisition systems that are heavily mechanical. Currently, on a daily basis, a seismic survey may require 800 or more crew members to place more than 200,000 prewired geophones over a field of several square miles. As such, the cost of cabling accounts for up to 50% of the total operating cost of a typical land survey, and up to 75% of the total equipment weight. This labor-intensive deployment of the prewired geophones, in addition to cost, prolongs the survey time and places a huge barrier on scaling the seismic acquisition and its adaptation/automation. Therefore, there has been a growing interest to switch from prewired geophones to wireless seismic acquisition. On the other hand, a typical seismic survey may generate tens of terabytes of raw seismic data per day. Hence, wireless communication faces great challenges in light of the enormous amounts of data that must be transmitted from geophones to on-site data collection centers.

Seismic Data Compression Using 2D Lifting-Wavelet Algorithms

Different seismic data compression algorithms have been developed in or-der to make the storage more efficient, and to reduce both the transmission time and cost. In general, those algorithms have three stages: transforma-tion, quantization and coding. The Wavelet transform is highly used tocompress seismic data, due to the capabilities of the Wavelets on representing geophysical events in seismic data. We selected the lifting scheme to implement the Wavelet transform because it reduces both computational and storage resources. This work aims to determine how the transforma-tion and the coding stages affect the data compression ratio.Several 2Dlifting-based algorithms were implemented to compress three different seis-mic data sets. Experimental results obtained for different filter type, filterlength, number of decomposition levels and coding scheme, are presented in this work.

Fast seismic data compression based on high‐efficiency SPIHT

Based on the set positioning in hierarchical tree (SPIHT) algorithm, fast seismic data compression based on the high-efficiency SPIHT (HESPIHT) algorithm is presented. To verify the validity of the algorithm, 10 seismic cross-sections are selected for the experiment. The result shows that the algorithm can reserve the quality of data better, meanwhile improving the efficiency of encoding and decoding.

Seismic data decomposition into spectral components using regularized nonstationary autoregression

Seismic data can be decomposed into nonstationary spectral components with smoothly variable frequencies and smoothly variable amplitudes. To estimate local frequencies, I use a nonstationary version of Prony’s spectral analysis method defined with the help of regularized nonstationary autoregression. To estimate local amplitudes of different components, I fit their sum to the data using regularized nonstationary regression. Shaping regularization ensures stability of the estimation process and provides controls on smoothness of the estimated parameters. Potential applications of the proposed technique include noise attenuation, seismic data compression, and seismic data regularization.

ON TIKHONOV REGULARIZATION AND COMPRESSIVE SENSING FOR SEISMIC SIGNAL PROCESSING

Using compressive sensing and sparse regularization, one can nearly completely reconstruct the input (sparse) signal using limited numbers of observations. At the same time, the reconstruction methods by compressing sensing and optimizing techniques overcome the obstacle of the number of sampling requirement of the Shannon/Nyquist sampling theorem. It is well known that seismic reflection signal may be sparse, sometimes and the number of sampling is insufficient for seismic surveys. So, the seismic signal reconstruction problem is ill-posed. Considering the ill-posed nature and the sparsity of seismic inverse problems, we study reconstruction of the wavefield and the reflection seismic signal by Tikhonov regularization and the compressive sensing. The l0, l1 and l2 regularization models are studied. Relationship between Tikhonov regularization and the compressive sensing is established. In particular, we introduce a general lp - lq (p, q ≥ 0) regularization model, which overcome the limitation on the assumption of convexity of the objective function. Interior point methods and projected gradient methods are studied. To show the potential for application of the regularized compressive sensing method, we perform both synthetic seismic signal and field data compression and restoration simulations using a proposed piecewise random sub-sampling. Numerical performance indicates that regularized compressive sensing is applicable for practical seismic imaging.