Robust dictionary learning for erratic noise corrupted seismic data reconstruction

• Autorzy: Zhao Qiang , Du Qizhen , Li Qingqing , Fu Liyun

Identyfikatory

DOI

10.1007/s11600-020-00433-0

• Języki publikacji: EN

Abstrakty

EN

Dictionary learning methods adaptively train their bases from the given data in an iterative manner; hence, they can capture more detailed features and achieve sparser representation than a method that uses a fxed basis. However, there also exists a good chance of erratic noise corrupting the dictionary because of the insufciency of the L1-norm regularization in distinguishing the signal and erratic noise, causing the inaccuracy of both dictionary learning and its application. A robust dictionary learning method is proposed to reconstruct the missing data even in the presence of strong erratic noise. Specifcally, a projected operator, which is constructed with the robust Huber misft, is used to damp erratic noise to a low-amplitude level. In the dictionary learning step, erratic noise is gradually reduced to an acceptable level with the help of this projected operator, from which the signal prototypes (or atoms) can be safely extracted in each iteration. Then, the learned dictionary is used to iteratively estimate the signal from the noisy and under-sampled data, performing noise attenuation as well as data interpolation at non-recording locations. We test the proposed dictionary learning method using 2D and 3D noisy seismic examples and compare it with other state-of-the-art methods. Numerical results demonstrate its superior efectiveness at recovering missing data and increasing signal-to-noise ratio in the presence of erratic noise.

Słowa kluczowe

EN

robust dictionary learning; seismic data reconstruction; erratic noise

PL

skuteczne uczenie się słownikowe; rekonstrukcja danych sejsmicznych; hałas nieregularny

• Wydawca: Instytut Geofizyki PAN ; Springer
• Czasopismo: Acta Geophysica
• Rocznik: 2020
• Tom: Vol. 68, no. 3
• Strony: 687--700
• Opis fizyczny: Bibliogr. 42 poz.

Twórcy

autor

Zhao Qiang

• Key Laboratory of Deep Oil and Gas, China University of Petroleum (East China), Qingdao 266580, China
• Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266237, China
• Key Laboratory of Geophysical Prospecting, CNPC, China University of Petroleum (East China), Qingdao 266580, China

autor

Du Qizhen

• Key Laboratory of Deep Oil and Gas, China University of Petroleum (East China), Qingdao 266580, China
• Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266237, China
• Key Laboratory of Geophysical Prospecting, CNPC, China University of Petroleum (East China), Qingdao 266580, China

autor

Li Qingqing

• Key Laboratory of Deep Oil and Gas, China University of Petroleum (East China), Qingdao 266580, China
• Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266237, China
• Key Laboratory of Geophysical Prospecting, CNPC, China University of Petroleum (East China), Qingdao 266580, China

autor

Fu Liyun

• Key Laboratory of Deep Oil and Gas, China University of Petroleum (East China), Qingdao 266580, China
• Laboratory for Marine Mineral Resources, Qingdao National Laboratory for Marine Science and Technology, Qingdao 266237, China
• Key Laboratory of Geophysical Prospecting, CNPC, China University of Petroleum (East China), Qingdao 266580, China

Bibliografia

• 1. Abma R, Kabir N (2006) 3D interpolation of irregular data with a POCS algorithm. Geophysics 71(6):E91–E97
• 2. Aharon M, Elad M, Bruckstein A (2006) K-SVD: an algorithm for designing over-complete dictionaries for sparse representation. IEEE Trans Signal Process 54(11):4311–4322
• 3. Aminzadeh F, Jean B, Kunz T (1997) 3-D salt and overthrust models. Society of Exploration Geophysicists
• 4. Beck A, Teboulle M (2009) A fast iterative shrinkage-thresholding algorithm for linear inverse problems. SIAM J Imaging Sci 2(1):183–202
• 5. Cai J-F, Ji H, Shen Z, Ye G-B (2014) Data-driven tight frame construction and image denoising. Appl Comput Harmonic Anal 37(1):89–105
• 6. Candes EJ, Li X, Ma Y, Wright J (2011) Robust principal component analysis? J ACM 58(3):11
• 7. Chen Y (2017) Fast dictionary learning for noise attenuation of multidimensional seismic data. Geophys J Int 209(1):21–31
• 8. Chen K, Sacchi MD (2014) Robust reduced-rank filtering for erratic seismic noise attenuation. Geophysics 80(1):V1–V11
• 9. Chen K, Sacchi MD (2017) Robust f − x projection filtering for simultaneous random and erratic seismic noise attenuation. Geophys Prospect 65(3):650–668
• 10. Chen Y, Ma J, Fomel S (2016a) Double-sparsity dictionary for seismic noise attenuation. Geophysics 81(2):V103–V116
• 11. Chen Y, Zhang D, Jin Z, Chen X, Zu S, Huang W, Gan S (2016b) Simultaneous denoising and reconstruction of 5-D seismic data via damped rank-reduction method. Geophys J Int 206(3):1695–1717
• 12. Claerbout JF, Muir F (1973) Robust modeling with erratic data. Geophysics 38(5):826–844
• 13. Combettes PL, Pesquet J-C (2007) A Douglas–Rachford splitting approach to non-smooth convex variational signal recovery. IEEE J Sel Top Signal Process 1(4):564–574
• 14. Donoho DL, Johnstone JM (1994) Ideal spatial adaptation by wavelet shrinkage. Biometrika 81(3):425–455
• 15. Elad M, Aharon M (2006) Image denoising via sparse and redundant representations over learned dictionaries. IEEE Trans Image Process 15(12):3736–3745
• 16. Gardner GH, Canning A (1994) Effects of irregular sampling on 3-D prestack migration. In: SEG technical program expanded abstracts 1994. Society of Exploration Geophysicists, pp 1553–1556
• 17. Guitton A, Symes WW (2003) Robust inversion of seismic data using the Huber norm. Geophysics 68(4):1310–1319
• 18. Herrmann FJ, Hennenfent G (2010) Non-parametric seismic data recovery with curvelet frames. Geophys J R Astron Soc 173(1):233–248
• 19. Herrmann FJ, Friedlander MP, Yilmaz O (2012) Fighting the curse of dimensionality: compressive sensing in exploration seismology. IEEE Signal Process Mag 29:88–100
• 20. Holland PW, Welsch RE (1977) Robust regression using iteratively reweighted least-squares. Commun Stat Theory Methods 6(9):813–827
• 21. Huber PJ (2011) Robust statistics. Springer, Berlin
• 22. Huber PJ, Ronchetti EM (2009) Robust statistics, 2nd edn. Wiley, Hoboken
• 23. Liang J, Ma J, Zhang X (2014) Seismic data restoration via data-driven tight frame. Geophysics 79(3):V65–V74
• 24. Liu L, Plonka G, Ma J (2017) Seismic data interpolation and denoising by learning a tensor tight frame. Inverse Prob 33(10):105011
• 25. Liu L, Ma J, Plonka G (2018) Sparse graph-regularized dictionary learning for suppressing random seismic noise. Geophysics 83(3):V215–V231
• 26. Mairal J, Bach F, Ponce J, Sapiro G (2009) Online dictionary learning for sparse coding. In: International conference on machine learning, ICML 2009, Montreal, Quebec, Canada, June, pp 689–696
• 27. Maronna R, Martin RD, Yohai V (2006) Robust statistics. Wiley, Chichester
• 28. Olshausen BA, Field DJ (1996) Emergence of simple-cell receptive field properties by learning a sparse code for natural images. Nature 381(6583):607–609
• 29. Oropeza V, Sacchi M (2011) Simultaneous seismic data denoising and reconstruction via multichannel singular spectrum analysis. Geophysics 76(3):V25–V32
• 30. Sacchi MD, Liu B (2005) Minimum weighted norm wavefield reconstruction for AVA imaging. Geophys Prospect 53(6):787–801
• 31. Sahoo SK, Makur A (2013) Dictionary training for sparse representation as generalization of k-means clustering. IEEE Signal Process Lett 20(6):587–590
• 32. Sahoo SK, Makur A (2015) Enhancing image denoising by controlling noise incursion in learned dictionaries. IEEE Signal Process Lett 22(8):1123–1126
• 33. Siahsar MAN, Gholtashi S, Kahoo AR, Chen W, Chen Y (2017) Data-driven multitask sparse dictionary learning for noise attenuation of 3D seismic data. Geophysics 82:V385–V396
• 34. Sternfels R, Viguier G, Gondoin R, Meur DL (2015) Multidimensional simultaneous random plus erratic noise attenuation and interpolation for seismic data by joint low-rank and sparse inversion. Geophysics 80(6):WD129–WD141
• 35. Trickett S, Burroughs L, Milton A (2012) Robust rank-reduction filtering for erratic noise. In: SEG technical program expanded abstracts 2012. Society of Exploration Geophysicists
• 36. Wang Y, Cao J, Yang C (2012) Recovery of seismic wavefields based on compres-sive sensing by an L1-norm constrained trust region method and the piecewise random subsampling. Geophys J Int 187(1):199–213
• 37. Wong RK, Lee T (2015) Matrix completion with noisy entries and outliers. arXiv preprint arXiv:1503.00214
• 38. Yu S, Ma J, Zhang X, Sacchi MD (2015) Interpolation and denoising of high-dimensional seismic data by learning a tight frame. Geophysics 80(5):V119–V132
• 39. Yu S, Ma J, Osher S (2016) Monte carlo data-driven tight frame for seismic data recovery. Geophysics 81(4):V327–V340
• 40. Zhao Q, Du Q (2017) Constrained data-driven tight frame for robust seismic data reconstruction. SEG technical program expanded, pp 4246–4250
• 41. Zhao Q, Du Q, Gong X, Chen Y (2018) Signal-preserving erratic noise attenuation via iterative robust sparsity-promoting filter. IEEE Trans Geosci Remote Sens 56(6):3547–3560
• 42. Zhao Q, Du Q, Sun W, Chen Y (2019) Iterative double Laplacianscaled low-rank optimization for under-sampled and noisy signalrecovery. IEEE Trans Geosci Remote Sens 57(11):9177–9187

Uwagi

Opracowanie rekordu ze środków MNiSW, umowa Nr 461252 w ramach programu "Społeczna odpowiedzialność nauki" - moduł: Popularyzacja nauki i promocja sportu (2021)