×

Discontinuity adaptive SAR image despeckling using curvelet-based BM3D technique. (English) Zbl 1457.94013

SAR images contain inherent multiplicative speckle noise caused by interference of transmitted signals with the returning ones. Speckle noise is visible as granular patterns and affects the image. Non-local means approches, like block matching and 3D filtering (BM3D), are quite effective for removing such kind of noise. BM3D exploits the self-similarities that occur in natural images to carry out noise filtering. This algorithm is suitalbe for removing additive noise but not necessarily for multiplicative speckle noise. So, in this paper an improvement of this method is proposed in order to remove this kind of noise from SAR images while preserving the curved edges and details. The improvement is three-fold. Firstly, the curvelet transform is applied in the place of the wavelet transform in the original algorithm. Secondly, importance sampling unscented Kalman filtering (ISUKF) is used instead of the Wiener filtering. And finally, the Manhattan distance measure is used to group similar patches instead of the Euclidean one. Each improvement is applied independently, giving more efficient algorithm each time. Finally, the algorithm with all these improvements gives better results in SAR image denoising than the state-of-the-arts methods.

MSC:

94A08 Image processing (compression, reconstruction, etc.) in information and communication theory
68U10 Computing methodologies for image processing

Software:

BM3D; NL-SAR
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] Gupta, A., Tripathi, A. and Bhateja, V., Despeckling of SAR images in contourlet domain using a new adaptive thresholding, \(3\) rd IEEE Int. Advance Computing Conf. (IACC) (2013), pp. 1257-1261.
[2] Azimi-Sadjadi, M. R. and Bannour, S., Two-dimensional adaptive block kalman filtering of SAR imagery, IEEE Trans. Geosci. Remote Sens.29(5) (1991) 742-753.
[3] Baraldi, A. and Panniggiani, F., A refined gamma MAP SAR speckle filter with improved geometrical adaptivity, IEEE Trans. Geosci. Remote Sens.33(5) (1995) 1245-1257.
[4] Candes, E., Demanet, L., Donoho, D. and Ying, L., Fast discrete curvelet transforms, Multiscale Model. Simul.5(3) (2006) 861-899. · Zbl 1122.65134
[5] Cha, S. H., Comprehensive survey on distance/similarity measures between probability density functions, City1(2) (2007) 1.
[6] Deledalle, C.-A., Denis, L. and Tupin, F., Iterative weighted maximum likelihood denoising with probabilistic patch-based weights, IEEE Trans. Image Process.18(12) (2009) 2661-2672. · Zbl 1371.94108
[7] Deledalle, C.-A., Denis, L. and Tupin, F., How to compare noisy patches? Patch similarity beyond Gaussian noise, Int. J. Comput. Vision99(1) (2012) 86-102. · Zbl 1254.68275
[8] Deledalle, C.-A., Denis, L., Tupin, F., Reigber, A. and Jager, M., NL-SAR: A unified nonlocal framework for resolution-preserving (Pol)(In)SAR denoising, IEEE Trans. Geosci. Remote Sens.53(4) (2015) 2021-2038.
[9] Cozzolino, D., Parrilli, S., Scarpa, G., Poggi, G. and Verdoliva, L., Fast adaptive nonlocal SAR despeckling, IEEE Geosci. Remote Sens. Lett.11(2) (2014) 524-528.
[10] Curlander, J. C. and McDonough, R. N., Synthetic Aperture Radar: Systems and Signal Processing (Wiley, 1991). · Zbl 0997.94518
[11] Dabov, K., Foi, A., Katkovnik, V. and Egiazarian, K., Image denoising by sparse 3D transform-domain collaborative filtering, IEEE Trans. Image Process.16(8) (2007) 2080-2095.
[12] K. Dabov, A. Foi, V. Katkovnik and K. Egiazarian, BM3D image denoising with shape-adaptive principal component analysis, SPARS’09-Signal Processing with Adaptive Sparse Structured Representations. 2009.
[13] Dai, L., Zhang, Y. and Li, Y., Image denoising using BM3D combining tetrolet prefiltering, Inf. Technol. J.12(10) (2013) 1995-2001.
[14] Dai, L., Zhang, Y. and Li, Y., BM3D image denoising algorithm with adaptive distance hard-threshold, Int. J. Signal Process. Image Process. Pattern Recognit.6(6) 41-50.
[15] Devapal, D., Kumar, S. S. and Jojy, C., Comprehensive survey on SAR despeckling techniques, Indian J. Sci. Technol.8(24) (2015).
[16] Devapal, D., Kumar, S. and Jojy, C., A novel approach of despeckling SAR images using nonlocal means filtering, J. Indian Soc. Remote Sens.45(3) (2016) 443-450.
[17] Frost, V. S., Stiles, J. A., Shanmugan, K. S. and Holtzman, J. C., A model for radar images and its application to adaptive digital filtering of multiplicative noise, IEEE Trans. Pattern Anal. Machine Intell.2 (1982) 157-166.
[18] Gleich, D. and Datcu, M., Wavelet-based despeckling of SAR images using Gauss-Markov random fields, IEEE Trans. Geosci. Remote Sens.45(12) (2007) 4127-4143.
[19] Guo, K. and Labate, D., Optimally sparse multidimensional representation using shearlets, SIAM J. Math. Anal.39(1) (2007) 298-318. · Zbl 1197.42017
[20] M. M. Hasan, Adaptive edge-guided block-matching and 3D filtering (BM3D) image denoising algorithm, thesis, The University of Western Ontario (2014).
[21] Jin, F., Fieguth, P., Winger, L. and Jernigan, E., Adaptive wiener filtering of noisy images and image sequences, in Proc. Int. Conf. Image Processing, ICIP, Vol. 3 (2003), pp. III-349.
[22] Jojy, C., Nair, M. S., Subrahmanyam, G. and Riji, R., Discontinuity adaptive non-local means with importance sampling unscented Kalman filter for de-speckling SAR images, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens.6(4) (2013) 1964-1970.
[23] Kotsakis, C. and Sideris, M. G., A modified Wiener-type filter for geodetic estimation problems with non-stationary noise, J. Geodesy75 (2001) 647-660. · Zbl 1049.86512
[24] Kuan, D. T., Sawchuk, A., Strand, T. C. and Chavel, P., Adaptive noise smoothing filter for images with signal dependent noise, IEEE Trans. Pattern Anal. Mach. Intel.7(2) (1985) 165-177.
[25] Lebrun, M., Image denoising method, Image Processing on Line2 (2012) 175-213.
[26] Lee, J. S. and Jong-Sen, Digital image enhancement and noise filtering by use of local statistics, IEEE Trans. Pattern Anal. Mach. Intell.2 (1980) 165-168.
[27] X. Li, H. Lu, G. Han and Z. Liang, A noise reduction method for non-stationary noise model of SPECT sinogram based on Kalman filter, Department of Radiology, State University of New York, Stony Brook, NY 11794, USA (IEEE, 2002).
[28] Lopes, A., Touzi, R. and Nezzy, E., Adaptive speckle filters and scene heterogeneity, IEEE Trans. Geosci. Remote Sens.28(6) (1990) 992-1000.
[29] J. Ma and G. Plonka, The curvelet transform, IEEE Signal Process. Mag.27(2) (2010) 118-133.
[30] Krishna Mohan, P. G., Santosh, Dadi Hari Hara and Surya Prasad, P., Improved denoising of SAR speckle images using curvelets over DT-CWT and traditional filters, Int. Conf. on RF and Signal Processing Systems, 2012.
[31] Parrilli, S., Poderico, M., Angelino, C. V. and Verdoliva, L., A nonlocal SAR image denoising algorithm based on LLMMSE wavelet shrinkage, IEEE Trans. Geosci. Remote Sens.50(2) (2012) 606-616.
[32] Pizurica, A., Philips, W. and Lemahieu, I., Despeckling SAR images using wavelets and a new class of adaptive shrinkage estimators, in Proc. Int. Conf. Image Process.2 (2001) 233-236.
[33] Ponomarenko, M., Pismenskova, M. and Egiazarian, K., Weighted MSE based spatially adaptive BM3D, in 25th European Signal Processing Conf. (EUSIPCO) (IEEE, 2017).
[34] Solbo, S. and Eltoft, T., Homomorphic wavelet-based statistical despeckling of SAR images, IEEE Trans. Geosci. Remote Sens.42(4) (2004) 711-721.
[35] Starck, J.-L., Candès, E. J. and Donoho, D. L., The curvelet transform for image denoising, IEEE Trans. Image Process.11(6) (2002) 670-684. · Zbl 1288.94011
[36] Subrahmanyam, G. R. S., Rajagopalan, A. and Aravind, R., A recursive filter for despeckling SAR images, IEEE Trans. Image Process.17(9) (2008) 1969-1974. · Zbl 1371.94356
[37] L. Torres and A. C. Frery, SAR Image Despeckling Algorithms using Stochastic Distances and Nonlocal Means, August 2013, preprint, arXiv:1308.4338.
[38] Zhong, H., Ma, K. and Zhou, Y., Modified BM3D algorithm for image denoising using nonlocal centralization prior, Signal Process.106 (2015) 342-347.
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.