Multiscale Methods Applied to the Analysis of Synthetic Aperture Radar Images
Méthodes Multiéchelles Appliquées à l'Analyse des Images Radar à Ouverture Synthétique
OPEN ACCESS
In this paper, we propose a filtering multiscale method to remove the speckle noise in synthetic aperture radar (SAR) images. This filtering is based on the à trous algorithm. It is derived from the multiscale methods developed for astronomical images using the extraction of significant structures. Nevertheless, the multiplicative behaviour of the speckle implies the wavelet thresholding to be modified according to the speckle noise statistic properties . We start with a classical approach based on a logarithmic transform of the image. Then, another method based on the energy of the image is presented . It allows one to obtain a better radiometrical precision in the filtered image . An original analysis is presented that takes advantage of the information given by the significant wavelet coefficients obtained from the thresholding procedure. This analysis is used to show the temporal variations at different scales and to extract the structures at small scales.
Résumé
Cet article propose une méthode de filtrage multiéchelle du bruit de speckle présent dans les images radar à ouverture synthétique. Ce filtrage est basé sur l'utilisation de l'algorithme à trous et s'inspire des méthodes multiéchelle d'extraction des structures significatives développées pour l'imagerie astronomique. Cependant, la nature multiplicative du bruit de speckle conduit à reconsidérer la méthode de seuillage dans l'espace des ondelettes et une première approche basée sur une transformation logarithmique de l'image est présentée. Une seconde approche, s'appuyant sur l'énergie du signal permet d'obtenir des images filtrées ayant une meilleure précision radiométrique . L'information fournie par les coefficients d'ondelettes significatifs est exploitée dans une analyse originale de l'image afin de mettre en évidence les variations temporelles des structures aux différentes échelles, et d'extraire les structures significatives aux petites échelles .
Multiscale Analysis, Wavelet Transform, Image Restoration, SAR images, Speckle .
Mots clés
Analyse multiéchelle, Ondelette, Restauration d'images, Imagerie ROS, Speckle .
[I] A. Bijaoui et F. Rué. A multiscale vision model adapted to the astronomical images. Signal Proc. v.46 n.3 pp. 345-362 1995.
[2] A. Bijaoui, J.L. Starck et F. Murtagh. Restauration des images multiéchelles parl'algorithme àtrous. Traitementdu Signal v.11 n3 pp. 229-243 1994.
[3] E Cauneau et T. Ranchin. Speckle removal inSAR images using the wavelet transform. Remote sensing for monitoring the changing environment of Europeed Winkler pp. 97-104, 1993.
[4] T.R. Crimmins. Geometric filter for speckle reduction.Appl. Opt. 24 pp.14381443, 1985.
[5] J.C. Dainty. Laser speckle and related phenomena. In Topics in Applied Physics,volume 9. ed. Springer-Verlag, Berlin, 1975.
[6] C. Elachi. Spaceborne Radar Remote Sensing :Applications and Techniques. IEEE Press, 1988.
[7] G. Franceschetti, V. Pascazio et G. Schirinzi. Iterative homomorphic technique for speckle reduction in synthetic aperture radar imaging. J.O.S.A. A 12 pp.686-694 1995.
[8] H. Guo, J.E. Odegard, M. Lang, R.A. Gopinath, I.Selesnick, et C.S. Burrus. Speckle reduction via wavelet shrinkage with application to SAR basedATD/R. In SPIE Math. Imaging : Wavelet Applications in Signal And Image Processing, pages 333-344, San Diego, CA, july 1994. volume 2303.
[9] W.W. Harman. Principles of the Statistical Theory of Communication, chapter 11, p. 217. Mac-Graw Hill, New York, 1963.
[10] A. Hillion et J.M.Boucher. Filtrage des images radar(SAR). Statistiques et analyse des données, 2 :35-57, 1991.
[11] D.H. Hoekman. Speckle ensemble statistics of logaritmically scaled data. IEEE transaction on geoscienee and remote sensing, 29 pp. 180-184,1991.
[12] M. Holdschneider, R. Kronland-Martinet, J. Morlet, et Ph. Tchamitchian. A real time algorithm for signal analysis with the help of the wavelet transform. Inwavelets,pages 286-297. ed. J.M. Combeset al. SpringerVerlag Berlin, 1989.
[13] M. Kendall et A. Stuart. The advanced theory of statistics, volume 2. Griffin, 1992.
[141 D.T. Kuan, A.A. Sawchuk, T.C. Strand and P. Chavel. Adaptativerestoration of images with speckle.IEEE Trans.A.S.S.P . 35 pp.373-383, 1987.
[15] P. Lascaux et R. Théodor, Analyse numérique matricielle appliquée àl'art de l'ingénieur, volume 2. Masson, 1994.
[16] J.S. Lee. Speckle analysis and smoothing of synthetic aperture radar images. C.G.I.P . 17 pp.24-32 1981.
[17] J.S. Lee. Digital image smoothing and the sigma filter. C.VG.I.P. 24 pp. 255-269 1983.
[18] J.S. Lee. Speckle suppression and analysis for synthetic aperture radar images. Proc.SPIE556 p. 170 1985.
[19] A. Lopès, E. Nezry, R. Touzi et H. Laur. Maximum a posteriori speckle filtering and first order texture models in SAR images. Proc. IGARSS'90 pp.2409-2412, 1990.
[20] S. Mallat. A Theory forMultiresolutionSignal Decomposition: The Wavelet Representation. IEEE Transactions on Pattern Analysis and Machine Intelligence., 11(7) :674-693, July 1989.
[21] D, Massonnet, M. Rossi, C. Carmona, F. Adragna, G. Peltzer, K. Feigl, et Th. Rabaute. The displacement field of the landers earthquake mapped by radar interferometry.Nature, 364(6433), 8 July 1993.
[22] F.Murtagh, J.L.Starck, et A.Bijaoui. Image restauration with noise suppression using the wavelet transform ii. Astron. and Astroph. Sup. See 112 pp. 179-189 1995.
[23] M. Saagi, J.M. Boucher,et G. Bénié. Hierarchical filtering of sar images. In IEEE IGARSS'95 Symposium, Florence, Italy, 10-14 July 1995.
[24] F. Safa et G. Flouzat. Speckle removal on radar imagery based on mathematical morphology. Signal processing, vol 16 no 2 :35-37, 1989.
[25] M. J. Shensa. The Discrete Wavelet Transform : Wedding the À Trous and Mallat Algorithms. IEEE Transactions on Signal Processing, 40(10) :2464-2482, October 1992.
[26] G. Strang. Wavelets and dilation equations : a brief introduction. SIAM Review, 31 :614-627, 1989.
[27] R. Touzi. Analyse d'images radar en télédétection : Améliorations radiométriques, ,filtrages du speckle et détection des contours. PhD thesis, Université Paul Sabatier de Toulouse, 1988.
[28] F.T. Ulaby, F. Kouyate, B. Brisco, et T.H. LeeWilliams.Textural information in SARimages. IEEE Transactions on Geoscience and Remote Sensing, GE-24(2), march 1986.
[29] M. Unser et A. Aldroubi. Polynomial splines and wavelets - a signal processing perspective. In Wavelets:a tutorial in theory and applications, pages 91-122. ed. C.K. Chui, Academic Press, New York, 1992.
[30] Y. Xu, J.B. Weaver, D.M. Healy, J. Lu. Wavelet transform domain filters a spatially selective noise filtration technique. IEEE Trans. Image Proc. v.3, pp. 747-758 1994.