S. Mallat. Zero-crossings of a wavelet transform. IEEE Trans. Inform. Theory, 37(4), July.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....shift sen sitivity that arises from downsamplers in the DWT implementation [4] Shift sensitivity is an undesirable property because it implies that DWT coefficients fail to distinguish between input signal shifts. Since downsamplers in the DWT implementation create shift sensitivity, Mallat [11], Beylkin [12] Coifman et al. 13] and Guo et al. 14] 15] devised the undecimated DWT, a wavelet trans form without downsamplers. Although the undecimated DWT is shift insensitive, it has high transform redundancy due to the absence of downsamplers. Unfortunately, the high transform redundancy ....

S. Mallat, "Zero-crossings of a wavelet transform," IEEE Trans. Inform. Theory, vol. 37, no. 4, July 1991.

.... with the changing scale is allowed) or immense computational complexity (e.g. the matchitg pursuit algorithm [40, 22] In some other methods, the resulting representations are non unique and involve approximate signal reconstructions, as is the case for zero crossing or local maxima methods [37, 30, 38, 39, 5]. Another approach has given up obtaining shift invariance and settled for a less restrictive property named shiftability [54, 3] which is accomplished by imposing limiting conditions on the scaling function [57, 3, 4] Recently, several authors proposed independently to extend the library of ....

S. Mallat, "Zero crossings of a wavelet transform", IEEE Trans. Inf. Theory, Vol. 37, No. 4, July 1991, pp. 1019 1033.

....proposed by Herley et al. 3] are sensitive to the signal location with respect to the chosen time origin. Shift invariant multiresolution representations exist. However, these methods either entail high oversampling rates [4] or alternatively, the resulting representations are non unique [5]. Mallat and Zhang [6] have suggested an adaptive matching pursuit algorithm. Under this proach the retainment of shift invariance, necessitates an oversized library containing the basis functions and all their shifted versions. The obvious drawbacks of matching pursuit are the rather high ....

S. Mallat, "Zero crossings of a wavelet transform", IEEE Trans. Inf. Th., Vol. 37, July 1991.

....representations exist. However, some methods either entail high oversampling rates (e.g. in [127, 9, 10, 86, 122] no down sampling with the changing scale is allowed) or alternatively, the resulting representations are non unique (as is the case for zero crossing or local maxima methods, e.g. [93, 74, 94, 95, 8]) Furthermore, zero crossing and related methods facilitate a signal reconstruction that is necessarily approximate. We also note that such methods lead to non orthogonal representations, rendering the interpretation of the correlation properties among the expansion coe#cients more di#cult. ....

.... with the changing scale is allowed) or immense computational complexity (e.g. the matching pursuit algorithm 23 [55, 96] In some other methods, the resulting representations are non unique and involve approximate signal reconstructions, as is the case for zero crossing or local maxima methods [8, 74, 93, 94, 95]. Another approach has relaxed the requirement for shift invariance, and defined a less restrictive property named shiftability [3, 134] which is accomplished by imposing limiting conditions on the scaling function [3, 6, 141] In this chapter, we define a shifted wavelet packet (SWP) library ....

S. Mallat, "Zero crossings of a wavelet transform", IEEE Trans. Inf. Theory, Vol. 37, No. 4, July 1991, pp. 1019--1033.

.... with the changing scale is allowed) or immense computational complexity (e.g. the matchitg pursuit algorithm [38, 23] In some other methods, the resulting representations are non unique and involve approximate signal reconstructions, as is the case for zero crossing or local maxima methods [35, 28, 36, 37, 3]. Another approach has given up obtaining shift invariance and settled for a less restrictive property named shiftability [46, 1] which is accomplished by imposing limiting conditions on the scaling function [49, 1, 2] Recently, several authors proposed independently to extend the library of ....

S. Mallat, "Zero crossings of a wavelet transform", IEEE Trans. Inf. Theory, Vol. 37, July 1991, pp. 1019 1033.

....of constant acceleration. Features include acceleration, velocity, arclength, order, and multiscale edgepoints. x(t) Hx1 Hx2 Hx3 Lx3 y(t) Hy1 Hy2 Hy3 Ly3 (a) b) c) y t y x t 3. 2 Object Trajectory Segmentation We implement the discrete wavelet transform as introduced by Mallat[6]. Since the mother wavelet is defined as the second order derivative of a smoothing function, the detailed coefficients estimate the acceleration of the trajectory at various resolutions. We combine acceleration measurements from the x and y trajectories to calculate the magnitude of the ....

Stephane Mallat, "Zero-Crossings of a wavelet transform," IEEE Transactions on Information Theory, Vol. 37, No. 4, pp. 1019-1033, July 1991.

.... MPEG 7 standard [1] Early work attempted to interpret images in terms of a semantic hierarchy [10] and relate these to quad trees [17] Proposals, for analyzing images into tree like hierarchies, from the signal and image processing community have included Laplacian pyramids [7,8] and wavelets [21, 22, 24]. Both are computationally efficient but they have some fundamental drawbacks in the context of producing trees: they have the potential to introduce spurious features at large scale (and therefore branches) large scale features are blurred and hence the resulting, large scale, fuzzy blobs do not ....

S. G. Mallat. Zero-crossings of a wavelet transform. IEEE Trans. on Information Theory, Vol. 11:pp 1019--1033, 1991.

.... i.e. in the non parametric identification of special signals, in the wavelet transform, and in image processing, ii) in matrix computation for the eigenvalue problem, iii) in the design of electrical and electronic simulation software tools for time domain and switching networks analysis [2, 3, 5, 4]. The problem consists in determining a time interval where a set of N devices (or mathematical models) work correctly. It happens while the characteristic function of all the N devices (or signals) are greater than zero. Mathematically the problem can be posed as: Given N real functions f i : x ....

Mallat., S., Zero Crossing of a Wavelet Transform. IEEE Trans. on Inf. Theory, Vol. 17, No. 4, 1019-1033. 1991.

....ace is constructed. The authors treat a digitized version of theproblem in theform of a recursive equation, followedultim(8E0 by a form ulation interm ofm;0;#(8EE# of error. We now turn our attention to wavelet basedmlet ds. The zero crossingsproblem in the one dim;L ; m case is studied in [11]; and theproblem involving theme(#F of m dulus of the gradient in one and two dim(8; S cases is studied in [12] In [11] the authoremho ys the location of the zeros of the wavelettransform at dyadic scales and the values of the integrals of the wavelettransform with consecutive zeros as ....

....by a form ulation interm ofm;0;#(8EE# of error. We now turn our attention to wavelet basedmlet ds. The zero crossingsproblem in the one dim;L ; m case is studied in [11] and theproblem involving theme(#F of m dulus of the gradient in one and two dim(8; S cases is studied in [12] In [11], the authoremho ys the location of the zeros of the wavelettransform at dyadic scales and the values of the integrals of the wavelettransform with consecutive zeros as thelimS of integration. The interpretation is that, when the wavelet is the second derivative of the sm othing function chosen, ....

[Article contains additional citation context not shown here]

S. Mallat, Zero-crossings of a wavelet transform, IEEE Trans. Inform. Theory, 37 (1991), pp. 1019--1033.

....h up to a boundary condition. Nevertheless, D. Marr calls the attention to the fact that perception involves more complex features. D. Marr s conjecture was proved by Hummel and Moniot, 7] with the additional hypothesis of having gradient information along the boundaries. Recently, S. Mallat, [11], used dyadic wavelets to compute the multiscale edges, and also to reconstruct an image from its edges. Mallat s method can be interpreted as a sampling and reconstruction scheme where the sampling points are the edges and the samples are the derivatives of the image in a proper scale. In spite ....

S. Mallat. Zero-crossings of a wavelet transform. IEEE Trans. Inform. Theory, pages 1019--1033, July 1991.

....other useful properties such as linearity and conservation of energy [8] 9] For practical implementations, CWT is computationally very complex. Dyadic Wavelet Transform (DWT) is the special case of CWT when the scale parameter is discretized along the dyadic grid ( j 2 ) j=1, 2. and b Z [10], i.e. DWT(f,j) 2 t t f f W j j Y = 3) where denotes convolution and ) 2 ( 2 1 ) 2 j j t t Y = Y . For an appropriately chosen wavelet, the wavelet transform modulus maxima denote the points of sharp variations of the signal [6] 10] 12] This property of DWT has been ....

....grid ( j 2 ) j=1, 2. and b Z [10] i.e. DWT(f,j) 2 t t f f W j j Y = 3) where denotes convolution and ) 2 ( 2 1 ) 2 j j t t Y = Y . For an appropriately chosen wavelet, the wavelet transform modulus maxima denote the points of sharp variations of the signal [6] [10] [12] This property of DWT has been proven very useful for detecting pitch periods of speech signals [11] An appropriately chosen wavelet is a wavelet that is the first derivative of a smooth function [6] Zero crossings of musical signals can be considered as points of sharp variation of the ....

Mallat, S., " Zero- crossing of a wavelet transform " , IEEE Trans. on Information Theory, 37(4): 1019-1033, (1991).

.... w# ### ) is equal to zero and w# ### w ### , i.e. no regularization is done on w ### . Other models may also be considered. When the image contains contours, it may be interesting to derive the model from the detected edge. Zerocrossing wavelet coe cients indicate where the edges are [24]. By averaging three wavelet coe cients in the direction of the detected edge, we get a value w # , from which we derive the SNR S # of the edge (S # 0 if there is no detected edge) The model value w# is set to w # if a contour is detected, and 0 otherwise. This approach has the ....

S.G. Mallat, Zero crossings of a wavelet transform, IEEE Trans. Inform. Theory 37 (4) (1991) 1019}1033.

....of multi scale zero crossing representation and the stability of the reconstruction. In onedimension case, under certain conditions, the signal can be reconstructed by the multi1scale zero crossing representation [11] and the reconstruction can be stabilized by some supplementary information [12]. For more information, we refer to [13] 1.3.3 Wavelet Maxima Algorithm Wavelet maxima algorithm reconstructs signals from the value and the position of the directional local maxim as of wavelet transform [14] 16] It was reported that for 256 Theta 256 Theta 8bits Lena, after 10 times ....

S. Mallat, "Zero-Crossing of a Wavelet Transform," IEEE Trans. Information Theory, vol. 37, no. 4, pp. 1019-1033, July 1991.

No context found.

S. Mallat. Zero-crossings of a wavelet transform. IEEE Trans. Inform. Theory, 37(4), July.

No context found.

S. G. Mallat. Zero-crossings of a wavelet transform. IEEE Trans. Inform. Theory, 37(4):1019--1033, July 1991.

No context found.

S. Mallat, "Zero-crossings of a wavelet transform, " IEEE Trans. Patt. Anal. Mach. Intell., vol. 37, no. 4, pp. 1019--1033, 1993.

No context found.

S. Mallat, "Zero-crossings of a wavelet transform, " IEEE Trans. Patt. Anal. Mach. Intell., vol. 37, no. 4, pp. 1019--1033, 1993.

No context found.

S. Mallat. Zero crossings of a wavelet transform. IEEE Trans. on Information Theory, 37(4):1019-- 33, July 1991.

No context found.

S. Mallat, Zero-crossing of a wavelets transform, IEEE Trans. Inf. Theory, 37(4):1019-1033, 1991.

No context found.

S. Mallat. Zero crossings of a wavelet transform. IEEE Trans. on Information Theory, 37(4):1019--33, July 1991.

No context found.

S. Mallat. Zero-crossings of a wavelet transform. IEEE Trans. Inform. Theory, 37(4), July 1991.

No context found.

S. Mallat, "Zero-crossings of a wavelet transform," IEEE Trans. Inform. Theory, vol. 37, pp. 1019--1033, July 1991.

No context found.

Mallat, S. Zero-Crossings of a Wavelet Transform, IEEE Trans. IT, Vol. 37, No. 4, 1019-1033, 1991.

No context found.

S. G. Mallat, "Zero crossings of a wavelet transform," IEEE Trans. Inf. Theory 37~4!, 1019--1033 ~1991!.

No context found.

S. G. Mallat, "Zero-crossings of a wavelet transform," IEEE Trans. Inform. Theory, vol. IT-37, pp. 1019-1033, July 1991.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC