CN111224938A - Wireless seismograph network compressed data transmission method - Google Patents

Wireless seismograph network compressed data transmission method Download PDF

Info

Publication number
CN111224938A
CN111224938A CN201911085869.XA CN201911085869A CN111224938A CN 111224938 A CN111224938 A CN 111224938A CN 201911085869 A CN201911085869 A CN 201911085869A CN 111224938 A CN111224938 A CN 111224938A
Authority
CN
China
Prior art keywords
data
seismic data
wireless
layer
seismic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201911085869.XA
Other languages
Chinese (zh)
Inventor
佟训乾
宾康成
张晓普
林君
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jilin University
Original Assignee
Jilin University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jilin University filed Critical Jilin University
Priority to CN201911085869.XA priority Critical patent/CN111224938A/en
Publication of CN111224938A publication Critical patent/CN111224938A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L69/00—Network arrangements, protocols or services independent of the application payload and not provided for in the other groups of this subclass
    • H04L69/04—Protocols for data compression, e.g. ROHC
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01V—GEOPHYSICS; GRAVITATIONAL MEASUREMENTS; DETECTING MASSES OR OBJECTS; TAGS
    • G01V1/00—Seismology; Seismic or acoustic prospecting or detecting
    • G01V1/22—Transmitting seismic signals to recording or processing apparatus
    • G01V1/223—Radioseismic systems
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geology (AREA)
  • Geophysics (AREA)
  • Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Acoustics & Sound (AREA)
  • Computer Security & Cryptography (AREA)
  • Geophysics And Detection Of Objects (AREA)

Abstract

The invention discloses a wireless seismograph network compressed data transmission method, which comprises the following steps: according to the correlation of the seismic data in the spatial domain, each wireless seismograph on the same measuring line uses a measuring matrix to encode the recorded seismic data and then completes relay data transmission, and therefore compression sensing of the seismic data in the spatial domain is completed. Specifically, when seismic data are observed in a spatial domain by using a compressed sensing theory, data acquired by each wireless seismograph is multiplied by corresponding column elements in a measurement matrix to complete data encoding. Then, the encoded data is summed in the multi-hop transmission process to complete the compression of the seismic data. A generative countermeasure network is used to simultaneously achieve a sparse representation of seismic data and a high precision reconstruction of the sparse coefficients. The new data transmission method not only can solve the problems of accumulated data transmission and energy consumption in the traditional method, but also can reduce the maximum data transmission quantity among the acquisition nodes.

Description

Wireless seismograph network compressed data transmission method
Technical Field
The invention relates to the field of wireless seismic data transmission method design, in particular to a wireless seismograph network compressed data transmission method.
Background
Seismic exploration is evolving towards large-scale and high-density acquisition. Wireless seismic exploration equipment has been widely used in recent years in the oil and gas industry due to its portability compared to conventional equipment with thick cables. In wireless seismic data acquisition, geophones are distributed in a measurement area and networked through wireless modules for multi-hop data transmission. This architecture is known as a Wireless Geophone Network (WGN), which is a special type of Wireless Sensor Network (WSN). WGNs differ from traditional WSNs in that thousands of geophones acquire data simultaneously at sampling rates on the order of KHz. Due to the large amount of data, data transmission is often limited by the bandwidth of the wireless channel, which makes real-time data collection a difficult task. The wireless seismograph horizontally supported by hardware of the existing acquisition system to realize large-scale seismic data acquisition has more problems. Therefore, the requirement of larger-scale seismic data acquisition work by improving the use efficiency of the existing hardware resources becomes a new research idea, and the design of an efficient wireless seismic data transmission method is the research background of the invention.
Disclosure of Invention
The invention aims to provide a wireless seismograph network compressed data transmission method, which is characterized in that each wireless seismograph on the same measuring line performs compressed sensing on seismic data recorded by the wireless seismograph in a spatial domain according to the correlation of seismic exploration data in the spatial domain, so that the data transmission quantity among the wireless seismographs is constant, the problem of unbalanced seismic data transmission is solved, and the maximum data transmission quantity among acquisition nodes is reduced.
In order to achieve the purpose, the invention provides the following scheme:
a wireless seismograph network compressed data transmission method comprises a training stage and a compressed acquisition stage, wherein in the training stage, at the beginning of seismic exploration data acquisition, a proper measurement matrix and a proper reconstruction mapping relation are determined according to priori knowledge to form a training sample set; in a compression acquisition stage, each wireless seismograph compresses and encodes seismic data according to a measurement matrix learned by a compression sensing algorithm based on a generated countermeasure network, and finally transmits a compressed measurement value to a data center; the data center recovers the received measurements into the original seismic data using the corresponding reconstructed mapping.
Further, the training sample set collects prior data capable of reflecting main geological features of the target detection region according to the characteristics of the exploration region, and the construction method comprises the following steps: directly selecting seismic exploration profile data of a similar area as a training sample; or establishing a plurality of approximate forward models according to the related geological data, and then generating seismic section data by the forward models to be used as training samples.
Furthermore, the compression coding comprises the step of carrying out compression coding based on a compression perception theory on each wireless seismograph in the data transmission process of the multi-hop network according to the measurement matrix obtained in the training stage after the seismic data are recorded.
Furthermore, in the process of acquiring the seismic data, the compression coding process of the multi-hop network comprises the steps that each wireless seismograph multiplies the recorded seismic data by the corresponding column elements in the measurement matrix, then, the data sent by the child wireless seismographs are added, and the obtained sum is sent to the father wireless seismograph.
Further, the seismic data recovery process comprises the steps that the data center firstly inputs the measured value received from the wireless gateway seismograph as a data reconstruction mapping relation obtained in a training stage, and the measured value after compressed sensing is recovered into original seismic data by using the learned reconstruction mapping relation.
Further, sparse representation of the seismic data and reconstruction of measured values are simultaneously completed through the generation countermeasure network, the generation countermeasure network is composed of a generator G and a discriminator D, the generator G and the discriminator D search an optimal solution by performing a countermeasure game, the discriminator D extracts the internal features of the seismic data by using two-dimensional convolution operation, and then sparsifying the convolution operation result by using a correction linear unit; the discriminator D totally adopts 3 layers of convolutional neural networks, each layer of convolutional neural network uses ReLU as an activation function, the first layer of convolutional neural network uses 32 convolutional kernels with the size of 7 multiplied by 7 to carry out the operation with the step length of 2; the second layer of convolutional neural network uses 128 convolutional kernels with the size of 5 × 5, the third layer of convolutional neural network uses 512 convolutional kernels with the size of 3 × 3, and the convolution operation steps of the second layer and the third layer are both 1.
Further, the generator G is composed of a transformation module GtAnd a recovery module GrThe method comprises the steps of constraining the solving process of reconstruction mapping relation by means of a transform domain learned by a discriminator D, carrying out flattening arrangement on input data b, then expressing the input data b by using a layer of fully-connected neural network with T multiplied by N neurons, and finally rearranging the output of the neural network to enable the output of the neural network to have the same size as original seismic data y, and fitting the mapping relation from data b to sparse s projection from detail characteristics to complement;
recovery module GrA module G for reciprocal operation with the discriminator D and recoveryrAnd 3 layers of deconvolution operations with the ReLU as an activation function are adopted, the deconvolution operation of the first layer corresponds to the convolution operation of the last layer in the discriminator D, and the deconvolution operations of the first layer and the last layer both adopt convolution kernels with the same size and the same number of convolution kernels. The analogy is continued until a recovery module GrThe last deconvolution operation in the middle layer corresponds to the first convolution operation in the discriminator D, and the sizes of convolution kernels of the two layers are the same.
Further, the wireless seismographs are located on the same survey line.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the correlation of the seismic data in the spatial domain, each wireless seismograph on the same measuring line uses the measuring matrix to encode the seismic data recorded by the wireless seismograph, and then relay type data transmission is completed, so that the compressed sensing of the seismic data in the spatial domain is completed. Specifically, when seismic data are observed once in a spatial domain by using a compressed sensing theory, data acquired by each wireless seismograph are multiplied by corresponding column elements in a measurement matrix to complete data encoding. Then, the encoded data is summed in the multi-hop transmission process. Unlike the summation operation in the traditional data transmission, the summation operation of the compressed sensing is a weighted sum, and the transmission load is balanced. When the data center receives the codes sent by the gateway seismograph, the measurement matrix completes one observation of a single measuring line, and one observation value is also called as one projection of seismic data. The number of data observations is determined by the dimensionality of the measurement matrix, with smaller dimensions giving higher data compression ratios. After all observations are made, the data collection center will obtain several projections of the observations. A generative countermeasure network is used to simultaneously achieve a sparse representation of the projections and a high precision reconstruction of the sparse coefficients. Therefore, the invention designs a compressed data transmission frame based on the multi-hop network according to the structural characteristics of the seismic exploration network, thereby realizing the high-efficiency data transmission with balanced load. The new data transmission method can solve the problems of accumulated data transmission and energy consumption in the conventional method.
The invention combines the observation process of compressed sensing with the network topology structure to design a compressed coding frame, and can effectively solve the problems of accumulated data transmission and energy consumption in the conventional seismic exploration. The basic structure of the countermeasure network is designed and generated according to the compressed sensing process, the problems of measurement matrix optimization and data recovery reconstruction are solved, the acquisition scale is enlarged while the energy consumption of data transmission of the whole measuring line is reduced, and the working efficiency of field seismic exploration is improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a schematic flow chart of a high-efficiency wireless seismograph network compressed data transmission method of the present invention;
FIG. 2 is a schematic diagram of a multi-hop network compression coding framework of an embodiment of a high-efficiency wireless seismograph network compression data transmission method of the present invention;
FIG. 3 is a schematic diagram of a multi-hop network compression encoding process of an embodiment of a high-efficiency wireless seismograph network compression data transmission method of the present invention;
FIG. 4 is a schematic diagram of a network structure of an arbiter D according to an embodiment of the present invention;
FIG. 5 shows a transformation module G of an embodiment of a method for transmitting compressed data of a wireless seismograph network according to the inventiontA network structure schematic diagram;
FIG. 6 shows a recovery module G according to an embodiment of the present inventionrA network structure schematic diagram;
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in fig. 1, the wireless seismograph network compressed data transmission method comprises a training sample set construction stage and a seismic data compression acquisition stage:
firstly, a training stage is required to be completed before the acquisition of seismic exploration data is started, and a proper measurement matrix and a reconstruction mapping relation are determined according to priori knowledge to form a training sample set, so that data support is provided for optimizing the measurement matrix and fitting the reconstruction mapping relation, and the measurement matrix and the reconstruction mapping relation are solved simultaneously. The method comprises the following steps of collecting prior data capable of reflecting main geological features of a target detection area according to the characteristics of the exploration area so as to ensure that a measurement matrix and data reconstruction mapping relation obtained according to a training sample set can be suitable for compression coding and data recovery in an acquisition stage, and the method can be mainly constructed by two methods: one method is to directly select seismic exploration profile data of similar areas as training samples; the other method is that a plurality of approximate forward models are established according to the relevant geological data, and then the seismic section data generated by the forward models are used as training samples;
secondly, after recording the seismic data, each wireless seismograph performs compressed sensing in the data transmission process of the multi-hop network according to the measurement matrix obtained in the training stage according to the data transmission framework shown in fig. 2.
And finally, inputting a measured value received by the data center from the wireless gateway seismograph as a data reconstruction mapping relation obtained in a training stage, and restoring the measured value subjected to compression coding into original seismic data by using the learned reconstruction mapping relation.
Referring to fig. 3, combining the observation process of compressive sensing with the network topology, and completing constant data transmission and data compression after data encoding, includes the following steps:
in the seismic data acquisition process, the compression encoding process of the multi-hop network is shown in fig. 3. Wherein, yjiFor a seismic data section y ∈ RT×NAcquiring seismic data of the ith wireless seismograph at the jth acquisition moment, wherein T is the acquisition time length, and N is the number of the wireless seismographs; phiilFor the measurement matrix phi epsilon RN×MRow i and column l. And each wireless seismograph multiplies the recorded seismic data by the corresponding row vector in the measurement matrix, then adds the data transmitted by the child wireless seismograph, and finally transmits the obtained sum to the father wireless seismograph.
With reference to fig. 3, 4, 5 and 6, the data compression coding transmission method simultaneously achieves the sparse representation of projection and the high-precision reconstruction of sparse coefficients by generating a countermeasure network, and the generator G and the discriminator D search for an optimal solution by performing a countermeasure game, including that the discriminator D first extracts the intrinsic characteristics of the seismic data by using two-dimensional convolution operation, and then thins the convolution operation result by using a modified linear unit (ReLU). A total of 3 layers of convolutional neural networks are adopted, and ReLU is used as an activation function in each layer of convolutional neural network. The first layer of convolutional neural network uses 32 convolutional kernels of size 7 × 7, and performs an operation with step size 2. Because the underlying convolutional neural network can extract some basic features of local details of the input data, the discriminator D uses a convolution kernel with a larger size in the convolution of the first layer to ensure that seismic data features with larger size and lower variation frequency can also be captured. The second layer of convolutional neural network uses 128 convolutional kernels of size 5 × 5, the third layer of convolutional neural network uses 512 convolutional kernels of size 3 × 3, and the convolution operation steps of the second and third layers are both 1. The second and third layers of convolutional neural networks in the discriminator D are mainly used for abstracting and combining the features extracted by the first layer of convolutional neural network, so that the generalization representation capability of the convolutional neural network is improved, and the network structure is shown in FIG. 4;
the generator G is divided into transformation modules GtAnd a recovery module GrConstraining the solution of the reconstructed mapping by means of the transform domain learned by the discriminator D, GtFirstly, input data b is subjected to flattening arrangement, then a layer of full-connection neural network with T multiplied by N neurons is used for representing the input data b, finally, the output of the neural network is rearranged to enable the output of the neural network to have the same size with the original seismic data y, the mapping relation from the data b to sparse s projection is fitted from detail characteristics for completion, and a transformation module G is used for completing the mapping relationtThe network structure of (2) is shown in fig. 5;
recovery module GrIs a decoder corresponding to the discriminator D, using 3 layers of deconvolution operation with ReLU as the activation function. The first layer of deconvolution operation corresponds to the last layer of convolution operation in the discriminator D, and both adopt convolution kernels with the same size and the same number of convolution kernels. The analogy is continued until a recovery module GrThe last deconvolution operation corresponds to the first convolution operation in the discriminator D, and the sizes of convolution kernels of the two convolution operations are bothThe same is true. Recovery module GrThe network structure of (2) is shown in fig. 6.
The principle and the implementation mode of the invention are explained by applying a specific example, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.

Claims (8)

1. A wireless seismograph network compressed data transmission method is characterized in that: the method comprises a training stage and a compression acquisition stage, wherein in the training stage, at the beginning of seismic exploration data acquisition, a proper measurement matrix and a reconstruction mapping relation are determined according to priori knowledge to form a training sample set; in a compression acquisition stage, each wireless seismograph compresses and encodes seismic data according to a measurement matrix learned by a compression sensing algorithm based on a generated countermeasure network, and finally transmits a compressed measurement value to a data center; the data center recovers the received measurements into the original seismic data using the corresponding reconstructed mapping.
2. The method of claim 1, wherein:
the training sample set collects prior data capable of reflecting main geological features of a target detection area according to the characteristics of the exploration area, and the construction method comprises the following steps: directly selecting seismic exploration profile data of a similar area as a training sample; or establishing a plurality of approximate forward models according to the related geological data, and then generating seismic section data by the forward models to be used as training samples.
3. The method of claim 1, wherein:
and the compressed coding comprises the step of carrying out compressed coding based on a compressed sensing theory in the data transmission process of the multi-hop network according to the measurement matrix obtained in the training stage after each wireless seismograph records the seismic data.
4. A method according to claim 3, characterized by: in the process of seismic data acquisition, the compression coding process of the multi-hop network comprises the steps that each wireless seismograph multiplies the recorded seismic data by the corresponding column elements in the measurement matrix, then, the data sent by the child wireless seismographs are added, and the obtained sum is sent to the father wireless seismograph.
5. The method of claim 1, wherein: the seismic data recovery process comprises the steps that the data center firstly inputs the measured value received from the wireless gateway seismograph as a data reconstruction mapping relation obtained in a training stage, and the measured value after compressed sensing is recovered into original seismic data by using the learned reconstruction mapping relation.
6. The method of claim 1, wherein the sparse representation of the seismic data and the reconstruction of the measured values are simultaneously accomplished by the generative confrontation network, which is composed of a generator G and a discriminator D, the generator G and the discriminator D search for the optimal solution by performing a confrontation game, the discriminator D first extracts the intrinsic features of the seismic data using a two-dimensional convolution operation, and then sparsifies the convolution operation results using a modified linear unit; the discriminator D totally adopts 3 layers of convolutional neural networks, each layer of convolutional neural network uses ReLU as an activation function, the first layer of convolutional neural network uses 32 convolutional kernels with the size of 7 multiplied by 7 to carry out the operation with the step length of 2; the second layer of convolutional neural network uses 128 convolutional kernels with the size of 5 × 5, the third layer of convolutional neural network uses 512 convolutional kernels with the size of 3 × 3, and the convolution operation steps of the second layer and the third layer are both 1.
7. The method of claim 6,
the generator G is composed of a transformation module GtAnd a recovery module GrFormed by remapping the transform domain learned by discriminator DConstraining the solving process of the shooting relation, firstly carrying out flattening arrangement on input data b, then expressing the input data b by using a layer of fully-connected neural network with T multiplied by N neurons, finally rearranging the output of the neural network to ensure that the output of the neural network has the same size with original seismic data y, and fitting the mapping relation from the data b to sparse s projection from detail characteristics for completion;
recovery module GrA module G for reciprocal operation with the discriminator D and recoveryrAnd 3 layers of deconvolution operations with the ReLU as an activation function are adopted, the deconvolution operation of the first layer corresponds to the convolution operation of the last layer in the discriminator D, and the deconvolution operations of the first layer and the last layer both adopt convolution kernels with the same size and the same number of convolution kernels. The analogy is continued until a recovery module GrThe last deconvolution operation in the middle layer corresponds to the first convolution operation in the discriminator D, and the sizes of convolution kernels of the two layers are the same.
8. A method as in claim 1 wherein the wireless seismographs are in the same line.
CN201911085869.XA 2019-11-08 2019-11-08 Wireless seismograph network compressed data transmission method Pending CN111224938A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201911085869.XA CN111224938A (en) 2019-11-08 2019-11-08 Wireless seismograph network compressed data transmission method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201911085869.XA CN111224938A (en) 2019-11-08 2019-11-08 Wireless seismograph network compressed data transmission method

Publications (1)

Publication Number Publication Date
CN111224938A true CN111224938A (en) 2020-06-02

Family

ID=70832691

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201911085869.XA Pending CN111224938A (en) 2019-11-08 2019-11-08 Wireless seismograph network compressed data transmission method

Country Status (1)

Country Link
CN (1) CN111224938A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112818762A (en) * 2021-01-15 2021-05-18 北华大学 Large-size composite material and rapid nondestructive testing method for sandwich structure thereof
CN113570493A (en) * 2021-07-26 2021-10-29 京东数科海益信息科技有限公司 Image generation method and device

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4509150A (en) * 1980-12-31 1985-04-02 Mobil Oil Corporation Linear prediction coding for compressing of seismic data
CN104345340A (en) * 2013-07-29 2015-02-11 中国石油化工股份有限公司 Jitter sparse sampling method in seismic data regularization
US9015007B2 (en) * 2010-12-13 2015-04-21 Southwest Research Institute Sensor array processor with multichannel reconstruction from random array sampling
CN107135004A (en) * 2017-04-20 2017-09-05 中国科学技术大学 A kind of adaptive real-time lossless compression method to earthquake data flow
CN107942377A (en) * 2018-01-05 2018-04-20 吉林大学 A kind of seismic data compression and reconstructing method
CN109490957A (en) * 2018-11-28 2019-03-19 华北电力大学 A kind of Reconstruction of seismic data method based on space constraint compressed sensing
CN110320555A (en) * 2019-06-28 2019-10-11 中国石油大学(华东) A kind of Reconstruction of seismic data method
CN110361778A (en) * 2019-07-04 2019-10-22 华北电力大学 A kind of Reconstruction of seismic data method based on generation confrontation network

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4509150A (en) * 1980-12-31 1985-04-02 Mobil Oil Corporation Linear prediction coding for compressing of seismic data
US9015007B2 (en) * 2010-12-13 2015-04-21 Southwest Research Institute Sensor array processor with multichannel reconstruction from random array sampling
CN104345340A (en) * 2013-07-29 2015-02-11 中国石油化工股份有限公司 Jitter sparse sampling method in seismic data regularization
CN107135004A (en) * 2017-04-20 2017-09-05 中国科学技术大学 A kind of adaptive real-time lossless compression method to earthquake data flow
CN107942377A (en) * 2018-01-05 2018-04-20 吉林大学 A kind of seismic data compression and reconstructing method
CN109490957A (en) * 2018-11-28 2019-03-19 华北电力大学 A kind of Reconstruction of seismic data method based on space constraint compressed sensing
CN110320555A (en) * 2019-06-28 2019-10-11 中国石油大学(华东) A kind of Reconstruction of seismic data method
CN110361778A (en) * 2019-07-04 2019-10-22 华北电力大学 A kind of Reconstruction of seismic data method based on generation confrontation network

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
XIAOPU ZHANG ETAL: "《An Efficient Seismic Data Acquisition Based on Compressed Sensing Architecture With Generative Adversarial Networks》", 《IEEE》 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112818762A (en) * 2021-01-15 2021-05-18 北华大学 Large-size composite material and rapid nondestructive testing method for sandwich structure thereof
CN112818762B (en) * 2021-01-15 2022-06-07 北华大学 Large-size composite material and rapid nondestructive testing method for sandwich structure thereof
CN113570493A (en) * 2021-07-26 2021-10-29 京东数科海益信息科技有限公司 Image generation method and device

Similar Documents

Publication Publication Date Title
CN107784676B (en) Compressed sensing measurement matrix optimization method and system based on automatic encoder network
CN110933429B (en) Video compression sensing and reconstruction method and device based on deep neural network
CN102148987B (en) Compressed sensing image reconstructing method based on prior model and 10 norms
CN107205154B (en) Wireless multimedia sensor network image compression acquisition method based on matrix completion
CN108960333A (en) Lossless compression method for high spectrum image based on deep learning
CN111224938A (en) Wireless seismograph network compressed data transmission method
CN112541572A (en) Residual oil distribution prediction method based on convolutional encoder-decoder network
Wang et al. Seismic velocity inversion transformer
Bin et al. Compressive data gathering with generative adversarial networks for wireless geophone networks
CN113359212A (en) Reservoir characteristic prediction method and model based on deep learning
CN113793261A (en) Spectrum reconstruction method based on 3D attention mechanism full-channel fusion network
CN111046737A (en) Efficient intelligent sensing acquisition method for microseism signal detection
Zeng et al. InversionNet3D: Efficient and scalable learning for 3-D full-waveform inversion
CN114418850A (en) Super-resolution reconstruction method with reference image and fusion image convolution
CN115185937A (en) SA-GAN architecture-based time sequence anomaly detection method
CN107942377A (en) A kind of seismic data compression and reconstructing method
CN108390871B (en) Radar data compression method based on autoregressive model frame prediction
CN117173022A (en) Remote sensing image super-resolution reconstruction method based on multipath fusion and attention
CN116863959A (en) Dolphin sound generating method based on generating countermeasure network
CN107622476A (en) Image Super-resolution processing method based on generative probabilistic model
CN110753224B (en) Data rearrangement method and system for measurement while drilling data compression
Liu et al. Diverse Hyperspectral Remote Sensing Image Synthesis With Diffusion Models
CN113053123B (en) Traffic prediction method and device based on space-time big data
CN111045861B (en) Sensor data recovery method based on deep neural network
Zhang et al. Missing measurement data recovery methods in structural health Monitoring: The State, challenges and case study

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
RJ01 Rejection of invention patent application after publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20200602