×

Steerable PCA for rotation-invariant image recognition. (English) Zbl 1326.92049

Summary: In this paper, we propose a continuous-domain version of principal-component analysis, with the constraint that the underlying family of templates appears at arbitrary orientations. We show that the corresponding principal components are steerable. Our method can be used for designing steerable filters so that they best approximate a given collection of reference templates. We apply this framework to the detection and classification of micrometer-sized particles that are used in a microfluidic diagnostics system. This is done in two steps. First, we decompose the particles into a small number of templates and compute their steerable principal components. Then we use these principal components to automatically estimate the orientation and the class of each particle.

MSC:

92C55 Biomedical imaging and signal processing
94A08 Image processing (compression, reconstruction, etc.) in information and communication theory
PDFBibTeX XMLCite
Full Text: DOI

References:

[1] W. T. Freeman and E. H. Adelson, {\it The design and use of steerable filters}, IEEE Trans. Pattern Anal. Mach. Intell., 13 (1991), pp. 891-906.
[2] Y. Hel-Or and P. C. Teo, {\it Canonical decomposition of steerable functions}, J. Math. Imaging Vision, 9 (1998), pp. 83-95. · Zbl 0906.68200
[3] S. Held, M. Storath, P. Massopust, and B. Forster, {\it Steerable wavelet frames based on the Riesz transform}, IEEE Trans. Image Process., 19 (2010), pp. 653-667. · Zbl 1371.42043
[4] R. Hilai and J. Rubinstein, {\it Recognition of rotated images by invariant Karhunen-Loeve expansion}, J. Opt. Soc. Amer. A, 11 (1994), pp. 1610-1618.
[5] M. Jacob and M. Unser, {\it Design of steerable filters for feature detection using canny-like criteria}, IEEE Trans. Pattern Anal. Mach. Intell., 26 (2004), pp. 1007-1019.
[6] M. Jogan, E. Zagar, and A. Leonardis, {\it Karhunen-Loeve expansion of a set of rotated templates}, IEEE Trans. Image Process., 12 (2003), pp. 817-825.
[7] R. Lenz, {\it Group-theoretical model of feature extraction}, J. Opt. Soc. Amer. A, 6 (1989), pp. 827-34.
[8] R. Lenz, {\it Group invariant pattern recognition}, Pattern Recognit., 23 (1990), pp. 199-217.
[9] A. Levy and J. Rubinstein, {\it Hilbert-space Karhunen-Loève transform with application to image analysis}, J. Opt. Soc. Amer. A, 16 (1999), pp. 28-35.
[10] M. Michaelis and G. Sommer, {\it A Lie group approach to steerable filters}, Pattern Recognit. Lett., 16 (1995), pp. 1165-1174.
[11] P. Perona, {\it Steerable-scalable kernels for edge detection and junction analysis}, Image Vis. Comput., 10 (1992), pp. 663-672.
[12] P. Perona, {\it Deformable kernels for early vision}, IEEE Trans. Pattern Anal. Mach. Intell., 17 (1995), pp. 488-499.
[13] C. Ponce and A. Singer, {\it Computing steerable principal components of a large set of images and their rotations}, IEEE Trans. Image Process., 20 (2011), pp. 3051-3062. · Zbl 1372.94211
[14] Z. Püspöki, C. Vonesch, and M. Unser, {\it Detection of symmetric junctions in biological images using \(2\)-D steerable wavelet transforms}, in Proceedings of the Tenth IEEE International Symposium on Biomedical Imaging: From Nano to Macro (ISBI ’13), 2013.
[15] E. P. Simoncelli and H. Farid, {\it Steerable wedge filters for local orientation analysis}, IEEE Trans. Image Process., 5 (1996), pp. 1377-1382.
[16] E. P. Simoncelli, W. T. Freeman, E. H. Adelson, and D. J. Heeger, {\it Shiftable multiscale transforms}, IEEE Trans. Inform. Theory, 38 (1992), pp. 587-607.
[17] G. Sommer, M. Michaelis, and R. Herpers, {\it The SVD approach for steerable filter design}, in Proceedings of the 1998 IEEE International Symposium on Circuits and Systems, Vol. 5, 1998, pp. 349-353.
[18] P. C. Teo and Y. Hel-Or, {\it LIE generators for computing steerable functions}, Pattern Recognit. Lett., 19 (1998), pp. 7-17. · Zbl 0907.68212
[19] M. Uenohara and T. Kanade, {\it Optimal approximation of uniformly rotated images: Relationship between Karhunen-Loeve expansion and discrete cosine transform}, IEEE Trans. Image Process., 7 (1998), pp. 116-119. · Zbl 0993.94526
[20] M. Unser and N. Chenouard, {\it A unifying parametric framework for \(2\) D steerable wavelet transforms}, SIAM J. Imaging Sci., 6 (2013), pp. 102-135. · Zbl 1279.68340
[21] C. Vonesch, F. Stauber, and M. Unser, {\it Design of steerable filters for the detection of micro-particles}, in Proceedings of the Tenth IEEE International Symposium on Biomedical Imaging: From Nano to Macro (ISBI ’13), 2013, pp. 934-937.
[22] Z. Zhao and A. Singer, {\it Fourier-Bessel rotational invariant eigenimages}, J. Opt. Soc. Amer. A, 30 (2013), pp. 871-877.
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.