Recently, deep learning methods have made a remarkable improvement in compressed sensing image recovery stage. In the compressed measurement stage, the existing methods measured by block by block owing to a huge measurement dictionary for the whole images and the high computational complexity. In this work, a novel deep convolutional neural network (DCNN) named Convolutional Measurement Compressed Sensing network (CMCS-net) is proposed for image compressed sensing considering both convolutional measurement (CM) and sparse prior. Different from existing works, the convolution operation is adopted both in the measurement phase and reconstruction phase, which retains the structure information of images much better. Simultaneously, the size of the measurement matrix is no longer limited by data dimensions. Particularly, by unfolding the CM process to analyse a Toeplitz-type matrix, the theoretical support of the convolutional compressed measurement is proposed. In addition, in the recovery phase, the authors consider the sparse prior in nature images by embedding the truncated hierarchical projection algorithm into their architecture to solve the problem of multilayered convolutional sparse coding. Furthermore, extensive experiments demonstrate that their proposed CMCS-net can marvellously reconstruct the images and fully remove the block artefact.

References

1. 1)
  - 14. Gu, S., Zuo, W., Xie, Q., et al: ‘Convolutional sparse coding for image super-resolution’. 2015 IEEE Int. Conf. on Computer Vision (ICCV), Santiago, Chile, 2015, pp. 1823–1831.
2. 2)
  - 13. Simon, D., Elad, M.: ‘Rethinking the CSC model for natural images’. Neural Information Processing Systems (NIPS), Vancouver, Canada, 2019, pp. 2274–2284.
3. 3)
  - 33. Fawzi, A., Davies, M.E., Frossard, P.: ‘Dictionary learning for fast classification based on soft-thresholding’, Int. J. Comput. Vis., 2014, 114, (2), pp. 306–321.
4. 4)
  - 22. Dong, W., Shi, G., Li, X., et al: ‘Image reconstruction with locally adaptive sparsity and nonlocal robust regularization’, Signal Process.-Image Commun., 2012, 27, (10), pp. 1109–1122.
5. 5)
  - 36. Lu, W., Li, W., Kpalma, K., et al: ‘Compressed sensing performance of random bernoulli matrices with high compression ratio’, IEEE Signal Process. Lett., 2015, 22, (8), pp. 1074–1078.
6. 6)
  - 25. Liu, C., Wang, J., Wang, W., et al: ‘Non-convex block-sparse compressed sensing with redundant dictionaries’, IET Signal Process., 2017, 11, (2), pp. 171–180.
7. 7)
  - 1. Rudin, L, Osher, S., Fatemi, E.: ‘Nonlinear total variation based noise removal algorithms’, Phys. D: Nonlinear Phenom., 1992, 60, pp. 259–268.
8. 8)
  - 39. Yuan, H., Song, H., Sun, X., et al: ‘Compressive sensing measurement matrix construction based on improved size compatible array LDPC code’, IET Image Process., 2015, 9, (11), pp. 993–1001.
9. 9)
  - 3. Li, C., Yin, W., Jiang, H., et al: ‘An efficient augmented lagrangian method with applications to total variation minimization’, Comput. Optim. Appl., 2013, 56, (3), pp. 507–530.
10. 10)
  - 24. Wang, H., Zhang, F., Wang, J., et al: ‘Estimating structural missing values via low-tubal-Rank tensor completion’. 2020 IEEE Int. Conf. on Acoustics, Speech and Signal Processing (ICASSP), Barcelona, Spain, 2020, pp. 3297–3301, doi: 10.1109/ICASSP40776.2020.9054531.
11. 11)
  - 31. Papyan, V., Sulam, J., Elad, M.: ‘Working locally thinking globally: theoretical guarantees for convolutional sparse coding’, IEEE Trans. Signal Process., 2017, 65, (21), pp. 5687–5701.
12. 12)
  - 15. Mallat, S.G., Zhang, Z.: ‘Matching pursuit with time-frequency dictionary’, IEEE Trans. Signal Process., 1993, 41, (12), pp. 3397–3415.
13. 13)
  - 18. Needell, N., Tropp, J.A.: ‘CoSaMP: iterative signal recovery from incomplete and inaccurate samples’, Appl. Comput. Harmon. Anal., 2009, 26, (3), pp. 301–321.
14. 14)
  - 8. Candes, E.J., Tao, T.: ‘Near-optimal signal recovery from random projections: universal encoding strategies?’, IEEE Trans. Inf. Theory, 2006, 52, (12), pp. 5406–5425.
15. 15)
  - 30. Yao, H., Dai, F., Zhang, S., et al: ‘DR2-net: deep residual reconstruction network for image compressive sensing’, Neurocomputing, 2019, 359, pp. 483–493.
16. 16)
  - 41. Gu, K., Li, L., Lu, H., et al: ‘A fast reliable image quality predictor by fusing micro- and macro-structures’, IEEE Trans. Ind. Electron., 2017, 64, (5), pp. 3903–3912.
17. 17)
  - 26. Wang, J., Huang, J., Zhang, F., et al: ‘Group sparse recovery in impulsive noise via alternating direction method of multipliers’, Appl. Comput. Harmon. Anal., 2020, 49, pp. 831–862, doi: 10.1016/j.acha.2019.04.002.
18. 18)
  - 35. Baraniuk, R.G., Davenport, M.A., Devore, R.A., et al: ‘A simple proof of the restricted isometry property for random matrices’, Constructive Approx., 2008, 28, (3), pp. 253–263.
19. 19)
  - 6. Candes, E.J., Wakin, M.B.: ‘An introduction to compressive sampling’, IEEE Signal Process. Mag., 2008, 25, (2), pp. 21–30.
20. 20)
  - 38. Haupt, J., Bajwa, W.U., Raz, G.M., et al: ‘Toeplitz compressed sensing matrices with applications to sparse channel estimation’, IEEE Trans. Inf. Theory, 2010, 56, (11), pp. 5862–5875.
21. 21)
  - 4. Buades, A., Coll, B., Morel, J., et al: ‘A non-local algorithm for image denoising’. 2005 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), San Diego, CA, USA, 2005, pp. 60–65.
22. 22)
  - 10. Zhang, F., Wang, J., Wang, W., et al: ‘Low-tubal-rank plus sparse tensor recovery with prior subspace information’, IEEE Trans. Pattern Anal. Mach. Intell. (T-PAMI), 2020, doi: 10.1109/TPAMI.2020.2986773.
23. 23)
  - 9. Yang, J., Wright, J., Huang, T.S., et al: ‘Image super-resolution via sparse representation’, IEEE Trans. Image Process., 2010, 19, (11), pp. 2861–2873.
24. 24)
  - 34. Candes, E.J., Romberg, J., Tao, T.: ‘Robust uncertainty principles: exact signal reconstruction from highly incomplete frequency information’, IEEE Trans. Inf. Theory, 2006, 52, (2), pp. 489–509.
25. 25)
  - 42. Deng, J., Dong, W., Socher, R., et al: ‘ImageNet: a large-scale hierarchical image databas’. 2009 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Miami, FL, USA, 2009, pp. 248–255.
26. 26)
  - 37. Zhang, G., Jiao, S., Xu, X.: ‘Compressed sensing and reconstruction with semi-hadamard matrices’. 2010 2nd Int. Conf. on Signal Processing Systems, Dalian, China, 2010.
27. 27)
  - 11. Papyan, V., Romano, Y., Sulam, J., et al: ‘Theoretical foundations of deep learning via sparse representations: a multilayer sparse model and its connection to convolutional neural networks’, IEEE Signal Process. Mag., 2018, 35, (4), pp. 72–89.
28. 28)
  - 12. Papyan, V., Romano, Y., Elad, M.: ‘Convolutional neural networks analyzed via convolutional sparse coding’, J. Mach. Learn. Res., 2017, 18, (1), pp. 2887–2938.
29. 29)
  - 28. Mousavi, A., Patel, A.B., Baraniuk, R.G.: ‘A deep learning approach to structured signal recovery’. 2015 53rd Annual Allerton Conf. on Communication, Control, Monticello, IL, USA, 2015, pp. 1336–1343.
30. 30)
  - 40. Zhou, X.: ‘Deep distributed convolutional neural networks: universality’, Anal. Appl., 2018, 16, (6), pp. 895–919.
31. 31)
  - 23. Huang, J., Zhang, T., Metaxas, D.N.: ‘Learning with structured sparsity’, J. Mach. Learn. Res., 2011, 12, pp. 3371–3412.
32. 32)
  - 5. Dabov, K., Foi, A., Katkovnik, V., et al: ‘Image denoising by sparse 3-D transform-domain collaborative filtering’, IEEE Trans. Image Process., 2007, 16, (8), pp. 2080–2095.
33. 33)
  - 17. Tropp, J.A., Gilbert, A.C.: ‘Signal recovery from random measurements via orthogonal matching pursuit’, IEEE Trans. Inf. Theory, 2007, 53, (12), pp. 4655–4666.
34. 34)
  - 2. Stanley, O., Martin, B., Donald, G., et al: ‘Aniterative regularization method for total variation-based image restoration’, Multiscale Model. Simul., 2005, 4, (2), pp. 460–489.
35. 35)
  - 7. Donoho, D.L.: ‘Compressed sensing’, IEEE Trans. Inf. Theory, 2006, 52, (4), pp. 1289–1306.
36. 36)
  - 20. Chen, S., Donoho, D.L., Saunders, M.A.: ‘Atomic decomposition by basis pursuit’, SIAM J. Sci. Comput., 1998, 20, (1), pp. 33–61.
37. 37)
  - 19. Khoramian, S.: ‘An iterative thresholding algorithm for linear inverse problems with multi-constraints and its applications’, Appl. Comput. Harmon. Anal., 2012, 32, (1), pp. 109–130.
38. 38)
  - 27. Dong, W., Shi, G., Li, X., et al: ‘Compressive sensing via nonlocal low-rank regularization’, IEEE Trans. Image Process., 2014, 23, (8), pp. 3618–3632.
39. 39)
  - 32. Xu, W., Wu, S., Er, M.J., et al: ‘New non-negative sparse feature learning approach for content-based image retrieval’, IET Image Process., 2017, 11, (9), pp. 724–733.
40. 40)
  - 16. Pati, Y.C., Rezaiifar, R., Krishnaprasad, P.S.: ‘Orthogonal matching pursuit: recursive function approximation with applications to wavelet decomposition’. Proc. of 27th Asilomar Conf. on Signals, Systems and Computers, Pacific Grove, CA, USA, 1993, pp. 40–44.
41. 41)
  - 21. Metzler, C.A., Maleki, A., Baraniuk, R.G.: ‘From denoising to compressed sensing’, IEEE Trans. Inf. Theory, 2016, 62, (9), pp. 5117–5144.
42. 42)
  - 29. Kulkarni, K., Lohit, S., Turaga, P., et al: ‘ReconNet: non-iterative reconstruction of images from compressively sensed measurements’. 2016 IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 2016, pp. 449–458.

CMCS-net: image compressed sensing with convolutional measurement via DCNN

References

Related content