Construction of Dense Time-Varying Array with Constraint Embedded Based on ControlVAE

SUN Wen-li; YUAN Da; JIANG Xin-bo

doi:10.12263/DZXB.20221144

您当前的位置：

首页 >

文章列表页 >

Construction of Dense Time-Varying Array with Constraint Embedded Based on ControlVAE

PAPERS | 更新时间：2026-04-10

- Construction of Dense Time-Varying Array with Constraint Embedded Based on ControlVAE
- ACTA ELECTRONICA SINICA Vol. 52, Issue 1, Pages: 354-363(2024)
- 作者机构：
  
  1.山东工商学院山东高校智能信息处理重点实验室，山东烟台 264005
  2.山东大学齐鲁交通学院，山东济南 250002
- 作者简介：
- 基金信息：
  
  Shandong Provincial Key Research and Development Program(2019GGX101040)
- DOI：10.12263/DZXB.20221144
  CLC： TP751.1;
- Received：11 October 2022，
  
  Revised：2023-05-25，
  
  Published：25 January 2024
- 稿件说明：
移动端阅览
孙文力,原达,姜新波.基于ControlVAE的约束嵌入稠密时变阵列构建[J].电子学报,2024,52(01):354-363.

SUN Wen-li,YUAN Da,JIANG Xin-bo.Construction of Dense Time-Varying Array with Constraint Embedded Based on ControlVAE[J].ACTA ELECTRONICA SINICA,2024,52(01):354-363.
孙文力,原达,姜新波.基于ControlVAE的约束嵌入稠密时变阵列构建[J].电子学报,2024,52(01):354-363. DOI： 10.12263/DZXB.20221144.

SUN Wen-li,YUAN Da,JIANG Xin-bo.Construction of Dense Time-Varying Array with Constraint Embedded Based on ControlVAE[J].ACTA ELECTRONICA SINICA,2024,52(01):354-363. DOI： 10.12263/DZXB.20221144.

摘要

在对探地雷达（Ground Penetrating Radar，GPR）数据进行三维建模与可视化分析过程中，其所依赖的三维时变阵列通常是由一维数据道或二维阵列间接生成的.由于采集到的数据往往比较稀疏，且存在不规则偏移，需要进行稠密处理，以获得高密度三维时变阵列.本文提出了一种基于可控变分自编码（Controllable Variational AutoEncoder，ControlVAE）的约束嵌入稠密时变阵列构建方法.使用基于ControlVAE的时变数据重构网络，通过隐空间采样插值和深浅层特征信息的融合，生成接近真实分布的伪道数据来增加阵列密度.构建了基于尺度不变特征变换的数据配准模块，提取道间梯度特征与结构对称性特征，可以在时空域内完成数据配准.同时将浅层特征作为约束信息嵌入数据重构网络，以消除数据道偏移对阵列生成的影响.实验结果表明，本文方法仅使用单测线二维B-scan数据集即可重建稠密时变阵列，可以有效降低重构模型累积误差，提高面对复杂真实数据时的鲁棒性.

Abstract

In the 3D modeling and visualization analysis of ground penetrating radar (GPR) data

the 3D time-varying arrays they rely on are usually generated indirectly from 1D data channels or 2D arrays. Since these data are often sparse and irregularly shifted

dense processing is required to obtain high-density 3D time-varying arrays. This paper proposes a constrained embedding dense time-varying array construction method based on controllable variational autoencoder (ControlVAE). A time-varying data reconstruction network based on the ControlVAE is used to increase the array density by generating pseudo-channel data close to the natural distribution through the fusion of hidden space sampling interpolation and deep and shallow feature information. A data alignment module based on scale-invariant feature transformation is constructed to extract the inter-channel gradient features and structural symmetry features to complete the data alignment in the spatial domain. The shallow features are also embedded as constraint information in the data generation network to eliminate the influence of data channel offset on array generation. The experimental results show that the proposed method can efficiently reconstruct a time-varying array with a single scan direction using a two-dimensional B-scan data set. This method can effectively reduce the cumulative error of the reconstructed model and improve the robustness in the face of complex actual data.

关键词

Keywords

references

WAI-LOK LAI W , DÉROBERT X , ANNAN P . A review of ground penetrating radar application in civil engineering: A 30-year journey from locating and testing to imaging and diagnosis [J]. NDT & E International , 2018 , 96 : 58 - 78 .

郝彤 , 赵杰 . 面向双曲线形态的探地雷达图像识别技术综述 [J]. 电子学报 , 2019 , 47 ( 6 ): 1366 - 1372 .

HAO T , ZHAO J . A brief review of the hyperbola signature recognition techniques for ground penetrating radar [J]. Acta Electronica Sinica , 2019 , 47 ( 6 ): 1366 - 1372 . (in Chinese)

WANG L X , YIN Y S , WANG H , et al . A method of reconstructing 3D model from 2D geological cross-section based on self-adaptive spatial sampling: A case study of cretaceous McMurray reservoirs in a block of Canada [J]. Petroleum Exploration and Development , 2021 , 48 ( 2 ): 407 - 420 .

DINH K , GUCUNSKI N , TRAN K , et al . Full-resolution 3D imaging for concrete structures with dual-polarization GPR [J]. Automation in Construction , 2021 , 125 : 103652 .

GHOSHAL S , BANU S , CHAKRABARTI A , et al . 3D reconstruction of spine image from 2D MRI slices along one axis [J]. IET Image Processing , 2020 , 14 ( 12 ): 2746 - 2755 .

YAMAGUCHI T , MIZUTANI T , NAGAYAMA T . Mapping subsurface utility pipes by 3-D convolutional neural network and Kirchhoff migration using GPR images [J]. IEEE Transactions on Geoscience and Remote Sensing , 2021 , 59 ( 8 ): 6525 - 6536 .

MA W P , WEN Z L , WU Y , et al . Remote sensing image registration with modified SIFT and enhanced feature matching [J]. IEEE Geoscience and Remote Sensing Letters , 2017 , 14 ( 1 ): 3 - 7 .

BERTOLAZZI E , FREGO M , BIRAL F . Point data reconstruction and smoothing using cubic splines and clusterization [J]. Mathematics and Computers in Simulation , 2020 , 176 : 36 - 56 .

ZHOU J E , JIA L , MENENTI M . Reconstruction of global MODIS NDVI time series: Performance of Harmonic ANalysis of Time Series (HANTS) [J]. Remote Sensing of Environment , 2015 , 163 : 217 - 228 .

LARI H H , NAGHIZADEH M , SACCHI M D , et al . Adaptive singular spectrum analysis for seismic denoising and interpolation [J]. Geophysics , 2019 , 84 ( 2 ): V133 - V142 .

LI X H , SHEN H F , ZHANG L P , et al . Sparse-based reconstruction of missing information in remote sensing images from spectral/temporal complementary information [J]. ISPRS Journal of Photogrammetry and Remote Sensing , 2015 , 106 : 1 - 15 .

VERLEYSEN C , DE VLEESCHOUWER C . Piecewise-planar 3D approximation from wide-baseline stereo [C]// 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE , 2016 : 3327 - 3336 .

YANG S D , YU X Y , ZHOU Y . LSTM and GRU neural network performance comparison study: Taking yelp review dataset as an example [C]// 2020 International Workshop on Electronic Communication and Artificial Intelligence (IWECAI) . Piscataway : IEEE , 2020 : 98 - 101 .

ZHANG H , GOODFELLOW I , METAXAS D , et al . Self-attention generative adversarial networks [EB/OL]. ( 2019-06-14 ) [ 2022-10-08 ]. https://arxiv.org/abs/1805.08318 https://arxiv.org/abs/1805.08318 .

KINGMA D P , WELLING M . Auto-encoding variational Bayes [EB/OL]. ( 2013-12-20 )[ 2022-10-08 ]. https://arxiv.org/abs/1312.6114 https://arxiv.org/abs/1312.6114 .

CRISTOVAO P , NAKADA H , TANIMURA Y , et al . Generating in-between images through learned latent space representation using variational autoencoders [J]. IEEE Access , 2020 , 8 : 149456 - 149467 .

DENTON E , BIRODKAR V . Unsupervised learning of disentangled representations from video [C]// Proceedings of the 31st International Conference on Neural Information Processing Systems . New York : ACM , 2017 : 4417 - 4426 .

SHAO H J , YAO S C , SUN D C , et al . ControlVAE: Controllable variational autoencoder [EB/OL]. ( 2020-04-13 ) [ 2022-10-08 ]. https://arxiv.org/abs/2004.05988 https://arxiv.org/abs/2004.05988 .

COMITE D , AHMAD F , AMIN M G , et al . Forward-looking ground-penetrating radar: Subsurface target imaging and detection: A review [J]. IEEE Geoscience and Remote Sensing Magazine , 2021 , 9 ( 4 ): 173 - 190 .

HOU S W , SUN W F , GUO B L , et al . RFSDAF: A new spatiotemporal fusion method robust to registration errors [J]. IEEE Transactions on Geoscience and Remote Sensing , 2022 , 60 : 1 - 18 .

WANG L G , WANG X Y , WANG Q M , et al . Investigating the influence of registration errors on the patch-based spatio-temporal fusion method [J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing , 2020 , 13 : 6291 - 6307 .

SHAO H J , XIAO Z S , YAO S C , et al . ControlVAE: Tuning, analytical properties, and performance analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2022 , 44 ( 12 ): 9285 - 9297 .

YANG X Y , ZHANG X L , ZHANG Z G , et al . DTWSSE: Data augmentation with a Siamese encoder for time series [C]// Asia-Pacific Web (APWeb) and Web-Age Information Management (WAIM) Joint International Conference on Web and Big Data . Cham : Springer , 2021 : 435 - 449 .

LI S , XU L , JING Y H , et al . High-quality vegetation index product generation: A review of NDVI time series reconstruction techniques [J]. International Journal of Applied Earth Observation and Geoinformation , 2021 , 105 : 102640 .

MIRZA M , OSINDERO S . Conditional generative adversarial nets [EB/OL]. ( 2014-11-06 )[ 2022-10-08 ]. https://arxiv.org/abs/1411.1784 https://arxiv.org/abs/1411.1784 .

HARFORD S , KARIM F , DARABI H . Generating adversarial samples on multivariate time series using variational autoencoders [J]. IEEE/CAA Journal of Automatica Sinica , 2021 , 8 ( 9 ): 1523 - 1538 .

ZHANG G , HUANG T C , GUERRERO T , et al . Use of three-dimensional (3D) optical flow method in mapping 3D anatomic structure and tumor contours across four-dimensional computed tomography data [J]. Journal of Applied Clinical Medical Physics , 2008 , 9 ( 1 ): 59 - 69 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Review on the Application of Metasurface-Based Impedance Matching Layer for Non-Destructive Testing Signal Enhancement

DABP: A Deep Learning Based Auto Focusing Back Projection Imaging Method for Ground Penetrating Radar

Suppression of Linear Interference Signal in GPR Profiling by Improved f-k Filtering Approach

Related Author

YUAN Da

LI Xiao-jing

ZHENG Wu-an

HAO Tong

LEI Wen-tai

LÜ Rong-qi

JIANG He-jun

SUI Hao

Related Institution

School of Physics Science and Engineering, Tongji University

Center for Spatial Information Science and Sustainable Development Applications, Tongji University

College of Surveying and Geo-Informatics, Tongji University

School of Electronic Information, Central South University

Science and Technology on Near-Surface Detection Laboratory

⁰