logo

SCIENTIA SINICA Informationis, Volume 48, Issue 8: 1051-1064(2018) https://doi.org/10.1360/N112017-00023

A novel 3-D reconstruction approach based on group sparsity of array InSAR

Hang LI1,2,3, Xingdong LIANG1,2,3,*, Fubo ZHANG1,2, Chibiao DING1,2,3, Yirong WU1,2,3
More info
  • ReceivedMar 10, 2017
  • AcceptedMay 17, 2017
  • PublishedJan 18, 2018

Abstract

As an advanced technology, synthetic aperture radar tomography (Tomo SAR) provides the feasibility to solve the layover problem caused by the inherent side-view mode in SAR sensors. However, the conventional nonparametric spectral estimation methods, such as truncated singular value decomposition (TSVD), are limited by the poor elevation resolution and cannot meet the needs of practical applications. Currently, SAR 3-D imaging methods based on compressive sensing are typically used; nevertheless, classic compressive sensing algorithms such as BP and OMP still exhibitproblems such as low efficiency, weak super-resolution performance, and poor anti-interference ability. Therefore, an algorithm with high robustness and super-resolution performance is significantly demanded.In this paper, a novel array InSAR 3-D reconstruction algorithm based on group-sparsity is proposed. It is improved based on the existing compressive sensing algorithm and exhibits better super-resolution and robustness. Based on the simulation data and the actual data of the first domestic 3-D imaging experiment by an airborne array InSAR, the super-resolution ability ofthe algorithm is verified, and the 3-D imaging results of the buildings are obtained.


Funded by


References

[1] Zhu X X, Bamler R. Space-borne high resolution tomographic interferometry. In: Proceedings of IEEE Geoscience and Remote Sensing Society (IGARSS), Cape Town, 2009. 869--872. Google Scholar

[2] Fornaro G, Serafino F. Imaging of Single and Double Scatterers in Urban Areas via SAR Tomography. IEEE Trans Geosci Remote Sens, 2006, 44: 3497-3505 CrossRef ADS Google Scholar

[3] Reigber A, Moreira A. First demonstration of airborne SAR tomography using multibaseline L-band data. IEEE Trans Geosci Remote Sens, 2000, 38: 2142-2152 CrossRef ADS Google Scholar

[4] Zhu X X, Bamler R. Tomographic SAR Inversion by $L_{1}$ -Norm Regularization-The Compressive Sensing Approach. IEEE Trans Geosci Remote Sens, 2010, 48: 3839-3846 CrossRef ADS Google Scholar

[5] Zhu X X, Bamler R. Super-Resolution Power and Robustness of Compressive Sensing for Spectral Estimation With Application to Spaceborne Tomographic SAR. IEEE Trans Geosci Remote Sens, 2012, 50: 247-258 CrossRef ADS Google Scholar

[6] Schmitt M, Stilla U. Compressive Sensing Based Layover Separation in Airborne Single-Pass Multi-Baseline InSAR Data. IEEE Geosci Remote Sens Lett, 2013, 10: 313-317 CrossRef ADS Google Scholar

[7] Yuan M, Lin Y. Model selection and estimation in regression with grouped variables. J R Statistical Soc B, 2006, 68: 49-67 CrossRef Google Scholar

[8] Akaike H. Information theory and an extension of the maximum likelihood principle. Inter Symp Inform Theory, 1992, 1: 610--624. Google Scholar

[9] Zhang F B. Research on signal processing of 3-D reconstruction in linear array synthetic aperture radar interferometry. Dissertation for Ph.D. Degree. Beijing: University of Chinese Academy of Sciences, 2015. Google Scholar

[10] Budillon A, Evangelista A, Schirinzi G. Three-Dimensional SAR Focusing From Multipass Signals Using Compressive Sampling. IEEE Trans Geosci Remote Sens, 2011, 49: 488-499 CrossRef ADS Google Scholar

[11] Zhang F B, Liang X D, Wu Y R. 3-D reconstruction for multi-channel SAR interferometry using terrain stagnation point based division. J Electron Inform Technol, 2015, 10: 2287--2293. Google Scholar

[12] Fornaro G, Lombardini F, Serafino F. Three-Dimensional Multipass SAR Focusing: Experiments With Long-Term Spaceborne Data. IEEE Trans Geosci Remote Sens, 2005, 43: 702-714 CrossRef ADS Google Scholar

[13] Tropp J, Gilbert A C. Signal recovery from random measurements via orthogonal matching pursuit. IEEE Trans Inf Theory, 2008, 53: 4655--4666. Google Scholar

[14] Chen S S, Donoho D L, Saunders M A. Atomic decomposition by basis pursuit. SIAM Rev, 2001, 43: 33--61. Google Scholar

[15] Donoho D L. Compressed sensing. IEEE Trans Inform Theor, 2006, 52: 1289-1306 CrossRef Google Scholar

[16] Huang J, Zhang T. The benefit of group sparsity. Ann Statist, 2010, 38: 1978-2004 CrossRef Google Scholar

[17] Zou H, Hastie T. Regularization and variable selection via the elastic net. J R Statistical Soc B, 2005, 67: 301-320 CrossRef Google Scholar

[18] Malioutov D, Cetin M, Willsky A S. A sparse signal reconstruction perspective for source localization with sensor arrays. IEEE Trans Signal Process, 2005, 53: 3010-3022 CrossRef ADS Google Scholar

[19] Tibshirani R, Johnstone I, Hastie T. Least angle regression. Ann Statist, 2004, 32: 407-499 CrossRef Google Scholar

  • Figure 1

    (a) The imaging geometry of array InSAR; (b) the layover model of array InSAR

  • Figure 2

    Workflow of the proposed approach of $3$-D reconstruction with array InSAR

  • Figure 3

    Comparison of scatters reconstruction with methods of TSVD, BP and our proposed algorithm. Interval of scatters is (a) 80 m, (b) 40 m, (c) 10 m, (d) 2 m, (e) 1 m, respectively; (f) magnification of part of (e)

  • Figure 4

    Performance evaluation of our proposed algorithm. Detection rate with different (a) super-resolution factors, (b) SNR, (c) array numbers; (d) detection rate with fixed $N~\times~$SNR; (e) super-resolution ability of different $N~\times~$SNR

  • Figure 5

    (a) Laying of corner reflectors with layover phenomenon; (b) SAR image of layover area

  • Figure 6

    Comparison of reconstruction of layover calibration pionts between methods of BP and our proposed algorithm. distance of 15 m with methods of (a) BP, (b) our proposed algorithm; distance of 1.8 m with methods of (c) BP, (d) our proposed algorithm

  • Figure 7

    (a) $2$-D SAR image of building areas; (b) result of $3$-D reconstruction with point clouds

  • Figure 8

    Result of $3$-D reconstruction of building areas with optical patch

  • 1   Table 1The system parameters of array InSAR
    Item Parameter Item Parameter
    Frequency 15 GHz Velocity 70 m/s
    Bandwidth 500 MHz Azimuth beam width 2$^{\circ}$
    PRF 1 kHz Range beam width 27$^{\circ}$
    Height 600 m Down-view angle 25$^{\circ}$

Copyright 2019 Science China Press Co., Ltd. 《中国科学》杂志社有限责任公司 版权所有

京ICP备18024590号-1