A Wavelet-Based Technique for Identifying, Labeling, and Tracking of Ocean Eddies

Size: px
Start display at page:

Download "A Wavelet-Based Technique for Identifying, Labeling, and Tracking of Ocean Eddies"

Transcription

1 MARCH 2002 LUO AND JAMESON 381 A Wavelet-Based Technique for Identifying, Labeling, and Tracking of Ocean Eddies JINGJIA LUO Department of Earth and Planetary Physics, Graduate School of Science, University of Tokyo, Tokyo, Japan LELAND JAMESON University of California, Lawrence Livermore National Laboratory, Livermore, California (Manuscript received 1 June 2000, in final form 20 July 2001) ABSTRACT Wavelet analysis offers a new approach for viewing and analyzing various large datasets by dividing information according to scale and location. Here a new method is presented that is designed to characterize timeevolving structures in large datasets from computer simulations and from observational data. An example of the use of this method to identify, classify, label, and track eddylike structures in a time-evolving dataset is presented. The initial target application is satellite data from the TOPEX/Poseiden satellite. But, the technique can certainly be used in any large dataset that might contain time-evolving or stationary structures. 1. Introduction Wavelet transforms provide information about a function or dataset with respect to scale and location in contrast to Fourier transforms, which provide a oneparameter family of coefficients representing the global frequency content (see Chui 1992; Daubechies 1988, 1992; Erlebacher 1996; Meyer 1990; Strang 1996). Due to their ability to resolve scales, wavelets can capture and identify local features of various structures contained in a given dataset that might be entirely missed by another form of analysis. In a word, we are searching for an efficient basis set with which to represent data that are local in nature. Such approaches have been carried out by others and one can see Saito (1998) for the search for an optimal basis set, Kirby (2001) for a complete discussion of basis sets of appropriate for various sets of data, and Jameson (2000) for wavelet analysis applied to numerical resolution of Kelvin Rossby waves. The idea is that wavelets form an efficient basis set for localized information such as ocean eddies. Our current application of wavelet analysis is to time-evolving structures found in oceanography. These structures might come from computational data or they might be contained in data collected by satellites. In either case, we are interested in separating meaningful structures such as eddies and fronts from noise and following the evolution of these structures with time. Corresponding author address: Dr. Leland Jameson, University of California, Lawrence Livermore National Laboratory, P.O. Box 808, MS L-312, Livermore, CA ameson3@llnl.gov In order to achieve our goal of completely identifying obects and structures that might appear in observational data or computational data, we begin here by building a straightforward database of wavelet signatures and from these base signatures we will have an idea of how to treat more complicated obects. Our signatures will be limited to those given in the Daubechies wavelet basis. We choose an orthogonal basis in order to have the ability to clearly separate information by scale and location without the possibility of overlap, which can happen in the case of a nonorthogonal bases sets. After choosing the wavelet, then we must specify ways to characterize information in datasets. One very simple and practical means of identifying and labeling obects in a dataset is by measuring the amount of energy that is contained at each scale. Simply put, one must define a region around an obect of interest, perhaps an eddy, and in this region one finds the energy present at each wavelet scale. This energy by scale approach can help to uniquely identify each structure, or specifically, each eddy. Noise is the structure that one can certainly expect to encounter in almost all data sources. So, it is a very important structure to understand. Once it is thoroughly understood in the wavelet basis, then it can more reliably be dealt with by, say, cancellation or other means. Our next relevant structure is a front, or near discontinuity in the dataset. The wavelet signature of a front will be easier to detect and identify than that of eddies and noise. For the case of a front we will see a sharp ump in the wavelet coefficients at most scales. In fact, if the front is a true discontinuity then the ump in the 2002 American Meteorological Society

2 382 JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY VOLUME 19 wavelet coefficients will be seen in all scales. Intuitively one would expect this since one way of viewing wavelet analysis is as a sequence of high-pass filters and a discontinuity will certainly have high-frequency energy at all scales. Our final task will be to follow and to track the various structures in our dataset as time evolves. 2. Wavelet analysis Possibly the most instructive way to think of wavelets is to compare with traditional analysis techniques such as Fourier analysis. With Fourier analysis we analyze discrete or continuous data using basis functions that are global, smooth, and periodic. This analysis yields a set of coefficients, say a k, which gives the amount of energy in the data at frequency k. Wavelet analysis, by contrast, analyzes data with basis functions that are local, slightly smooth, not periodic, and which vary with respect to scale and location. Wavelet analysis thereby produces a set of coefficients b,k which give the amount of energy in the data at scale and location k. Wavelet analysis can serve as a good complement to Fourier analysis. In fact, data that are efficiently analyzed with Fourier analysis often is not efficiently analyzed with wavelet analysis and the opposite situation also holds. For our purposes here we will confine our discussion to the so-called orthogonal wavelets and specifically the Daubechies family of wavelets. The orthogonality property leads to a clear indication when data deviate from a low-order polynomial, the importance of which will become clear when we discuss numerical methods. a. Defining the Daubechies wavelet To define Daubechies-based wavelets, see Daubechies (1988) and Erlebacher (1996), consider the two functions (x), the scaling function, and (x), the wavelet. The properties of the scaling function and wavelet are defined in terms of the low-pass filter coefficients h k and the high-pass filter coefficients g k. Here g k is actually the same filter as h k with the order reversed and with an alternating sign on the coefficients. More precise details on these high- and low-pass filters will be given in sections 2b and 2d after our current and more general introduction. First we define the dilation equation by L1 k k0 (x) 2 h (2x k), (1) which carries the name dilation equation since the independent variable x appears alone on the left-hand side but is multiplied by 2, or dilated, on the right-hand side. One also requires that the scaling function (x) be normalized: # (x) dx 1. The wavelet (x) is defined in terms of the scaling function, L1 k k0 (x) 2 g (2x k). (2) One builds an orthonormal basis from (x) and (x) by dilating and translating to get the following functions: /2 k(x) 2 (2 x k), and (3) /2 k(x) 2 (2 x k), (4) where, k Z. Here is the dilation parameter and k is the translation parameter. b. The spaces spanned by wavelets It is usual to let the spaces spanned by k (x) and k (x) over the parameter k, with fixed, be denoted by V and W, respectively. span V k Z k(x), (5) span k Z k W (x). (6) The scaling function subspaces form a nested set according to scale and the nesting is V1 V0 V1 (7) The wavelet subspaces W are the difference between the scaling function subspaces at different scales. The spaces V and W are related by V V1 W 1, (8) where the notation V 0 V 1 W 1 indicates that the vectors in V 1 are orthogonal to the vectors in W 1 and the space V 0 is simply decomposed into these two component subspaces. c. The high- and low-pass filters and orthogonality L1 L1 The coefficients H {h k} k0 and G {g k} k0 are related by g k (1) k h Lk for k 0,...,L 1. All wavelet properties, hence the definition of the wavelet being used, are specified through the parameters H and G. If one s data are defined on a continuous domain such as f(x) where x R is a real number, then one uses k(x) and k(x) to perform the wavelet analysis. If, on the other hand, one s data are defined on a discrete domain such as f(i) where i Z is an integer then the data are analyzed, or filtered, with the coefficients H and G. In either case, the scaling function (x) and its defining coefficients H detect localized low-frequency information; that is, they are low-pass filters, and the wavelet (x) and its defining coefficients G detect localized high-frequency information; that is, they are high-pass filters. Specifically, H and G are chosen so that dilations and translations of the wavelet, k (x), form an orthonormal basis L 2 (R) and so that (x) has M vanishing moments that determine the accuracy of the wavelet approximation. That is, scaling functions have the ability to represent polynomials up to a given order exactly and wavelets are, therefore, orthogonal to these same polynomials. This orthogonality is in terms of vanishing moments and therefore the number of vanishing

3 MARCH 2002 LUO AND JAMESON 383 moments determines the accuracy of the approximation. In other words, (x) will satisfy k m klm k(x) l (x) dx, (9) where kl is the Kronecker delta function, and the accuracy is specified by requiring that (x) (x) 0 satisfy 0 m (x)x dx 0, (10) for m 0,..., M 1. Under the conditions of the previous two equations, for any function f(x) L 2 (R) there exists a set {d k } such that where k f (x) d k(x), (11) Z k Z dk f (x) k(x) dx. (12) d. Quadrature mirror filters and the Haar wavelet The two sets of coefficients H and G are known as quadrature mirror filters. For Daubechies wavelets the number of coefficients in H and G, or the length of the filters H and G, denoted by L, is related to the number of vanishing moments M by 2M L. For example, the famous Haar wavelet is found by defining H as h 0 h 1 1. For this filter, H, the solution to the dilation equation (1), (x), is the box function: (x) 1 for x [0, 1] and (x) 0 otherwise. The Haar function is very useful as a learning tool, but because of its low order of approximation accuracy and lack of differentiability it is of limited use as a basis set. The coefficients H needed to define compactly supported wavelets with a higher degree of regularity can be found in Daubechies (1988). As is expected, the regularity increases with the support of the wavelet. Note that the usual notation to denote a Daubechies-based wavelet defined by coefficients H of length L is D L D 2M. However, one can also find examples where the Daubechies wavelet is denoted by a subscript that is the number of vanishing moments D M. We will use the first and most common convention for notation. e. Setting a largest and smallest scale In a continuous wavelet expansion, functions will arbitrarily small-scale structures can be represented. In practice, however, there is a limit to how small the smallest structure can be depending on, for example, the numerical grid resolution or the sampling frequency in a signal-processing scenario. Hence, on a computer an approximation, using Eq. (8), would be constructed in a finite space such as V W W W V, J J with the approximation being with J J J V0 k k k k k Z 1 k Z P f(x) s (x) d (x), (13) J J k k k k d f (x) (x) dx, s f (x) (x) dx utilizing orthogonality. Within this expansion, the scale 0 is arbitrarily chosen as the finest scale required, and scale J would be the scale at which a kind of local J average, k (x), provides sufficient large-scale information; that is, the first term in Eq. (13) provides the local mean around which the function oscillates. One must also limit the range of the location parameter, k. Assuming periodicity of f(x) implies periodicity on all wavelet coefficients, sk and dk, with respect to k. For the nonperiodic case, since k is directly related to the location, a limit is imposed on the values of k when the location being addressed extends beyond the boundaries of the domain. f. Implementation on a computer The wavelet decomposition matrix is the matrix embodiment of the dilation equation, Eq. (1), defining the scaling function and the accompanying equation, Eq. (2), defining the wavelet. The following two recurrence relations for the coefficients, sk and dk, in Eq. (13) are given as L 1 k n n2k2 n1 L 1 k n n2k2 n1 s hs and (14) d gs (15) as obtained from Eqs. (1) and (2), and we recall that h n refers to the chosen filter while we have g n (1) n h Ln. Denote the decomposition matrix embodied by these,1 two equations, assuming periodicity, by P N where the matrix subscript denotes the size of the square matrix while the superscripts indicate that P is decomposing from scaling function coefficients at scale to scaling function and wavelet function coefficients at scale,1 1; that is, maps s onto s 1 and d 1 : P N [ ] d s 1,1 P N : [s ], (16) 1 where we by s refer to the vector containing the coefficients at scale. Note that the vectors at scale 1 are half as long as the vectors at scale.

4 384 JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY VOLUME 19 FIG. 1. Contour plots for a Gaussian eddy and the same eddy with computer-generated noise superimposed. g. Wavelet analysis in two dimensions Performing the wavelet analysis of a given field in higher dimensions is a straightforward extension of the ideas from one dimension. Once can simply perform the wavelet analysis in a tensor product approach dimension by dimension. In fact, the regions of the domain which can be felt, as discussed above, carry over directly dimension by dimension. The tensor product approach works as follows: Recall that in one dimension a decomposition of the finest scale subspace yields V0 V1 W 1. So, if one takes the tensor product of this one-dimensional analysis with another one-dimensional analysis then one obtains V0 V0 (V1 W 1) (V1 W 1), (17) which yields V0 V0 (V1 V 1) (V1 W 1) One can think of the subspace V1 V1 (W V ) (W W ). (18) as representing the horizontal and vertical average, lowpass filtering, of the information contained in V 0 V 0. The subspace V1 W1 represents a horizontal low-pass filtering process and a vertical high-pass filtering process. Such a subspace would capture horizontal edges. On the other hand, the subspace W1 V1 would represent a horizontal high-pass filtering process and a vertical low-pass filtering process. One would be able to detect vertical edges in such a subspace. Finally, when the high-pass filter is applied in both the vertical and horizontal directions one arrives at the subspace, W1 W 1. It is this subspace that we are primarily interested in since it will detect variation in both the horizontal and vertical directions. Note that as in the one-dimensional case we certainly will perform many levels of wavelet decompositions, and the subspaces that we are primarily interested in will be W W, for 1, 2, 3, 4... Generally, decompositions up to 4 will be sufficient. 3. Defining wavelet signatures In this section we document the wavelet signatures, see below for definition of signatures, of various structures that will commonly appear in the wavelet analysis of various fields associated with oceanography. Examples of such structures would be noise, eddies, fronts, etc. In order to achieve our goal of completely identifying obects and structures that might appear in observational data or computational data, we begin here by building a straightforward database of wavelet signatures. For now our signatures will be limited to those given in the Daubechies wavelet basis. a. Various ways to form a wavelet signature Our first task is a brief discussion of the various ways that one could form a signature using a wavelet basis. In a word, wavelets give us information with respect to scale and location. One could imagine a very large num-

5 MARCH 2002 LUO AND JAMESON 385 ber of ways to use this information to form a uniquely identifying signature. We will explore only a few of these ways. b. Local variance or energy as a signature One possible way to find a wavelet signature is to find the amount of energy present at the various wavelet scales in a given region. This could be accomplished by summing up the square of the wavelet coefficients at the various scales in the following manner: 1 2 K,K,K C, (19) where will be the wavelet-detected variation at scale and C will be a corresponding scaling constant. Here is defined as k1k1n k2k2n k3k3n (2n 1) k1k1n k2k2n k3k3n [d (k, k, k )]. (20) As above, the parameter n will define a box in three dimensions about which the summation of the wavelet coefficients occurs. The reason that one would have a different n for each scale is that the wavelets at the larger scales, higher values of, will cover larger portions of the domain. One can see this from the above discussion of regions of influence. Therefore, one would expect that the values of n will decrease as increases. This will, of course, depend on the size of the region that one wished to use in finding their estimate of the energy at a given wavelet scale. The point (K 1, K 2, K 3 ) will be a wavelet translation index at the scale corresponding to a point in the physical space. The point is roughly centered at the location of the physical space where one needs an estimate of variance or energy. c. The wavelet subspaces for signature measurement We begin by observing the energy signature at various wavelet scales using the D4 wavelet. Recall from our earlier discussion that a wavelet decomposition in two dimensions provides us with four subspaces, V V, V W, W V, and W W. For now, we will give the energy contained in the subspaces W W, for 1, 2, 3,.... This energy will simply be the summation of the squares of the wavelet coefficients at scale in both the horizontal and vertical directions. These energies will give us a kind of wavelet noise energy spectrum at various scales. Note that in detecting the global energy in a domain one is not using the key strength of wavelet analysis, that is, the ability to separate structures by location. That is, one key difference between wavelet analysis and Fourier analysis is the information on location or the parameter k in the coefficient. In taking a global sum, this information d k FIG. 2. (a) The energy profile of noise with most of the energy in the smaller scales. (b) The energy profile of the single Gaussian eddy with noise added. is not used. However, this global analysis is an essential first step in building our wavelet signature database. d. Wavelet spectra and a bounding box Note that the manner in which one uses the information from wavelet analysis is far from unique and that in some manner one must combine the obtained information into some kind of meaningful and easily utilized information. For our purposes we have chosen to combine the wavelet information by scale and this has worked quite effectively. Further, we have a bounding box defined around each eddy for practical convenience. Without such a clearly defined bounding box one can encounter practical problems such as deciphering where one eddy ends and where the next eddy begins. Of course, deciding the beginning and end of an eddy becomes at the most precise level somewhat arbitrary, one still must have a precise manner to separate information. For our purposes we have not found the bounding box to be a limitation. If, however, situ-

6 386 JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY VOLUME 19 FIG. 3. The contour diagram of two eddies of different strength and size. ations are encountered in the future where the bounding box is a limitation, then appropriate modifications can be incorporated. 4. A catalog of wavelet signatures In this section we conduct testing of the technique on relatively simple datasets that consist of Gaussian structures called Gaussian eddies in this study. Of course, one could spend a great deal of time building models of eddies more realistic than ours, but as one will see, our Gaussian eddies are sufficient for testing our method, which will be proven by effectiveness when applied to real TOPEX/Poseidon (T/P) datasets. a. One Gaussian eddy with and without noise In this subsection we begin with our simplest test case, that of a single Gaussian eddy first without noise and then with computer-generated noise. In Fig. 1 we have two contour plots of sea surface height which indicates the size and strength of the eddy. The left plot is of the pure eddy and the right plot we show the eddy with computer-generated noise superimposed on it. This eddy is very simple but a good place to begin our catalog. In Fig. 2a we see the wavelet energy spectrum of the noise alone, and then in Fig. 2b the eddy alone and finally the eddy plus the noise. The horizontal axis represents a measure of energy in dimensionless units. The vertical axis is the wavelet scale FIG. 4. The wavelet energy profiles of two eddies of different strength and size. with the smaller numbers representing the smaller scales. Essentially, there are two features that should be noted: first, the energy level of the noise is much lower than the energy level of the eddy. Of course, since the noise is computer generated, we could have made it as strong as we like. However, we tried to make it as close as possible to noise levels found in real datasets. Second, note the structure of the energy spectrum for the noise and the eddy. One can see that the noise spectrum has most of its energy in the smallest scales and the energy level decreases as the scale size goes up. On the other hand, the energy spectrum of the eddy is quite different. The eddy has more and more energy with increasing scale until it reaches a scale 6 at which point the energy decreases. The values of the energy at scale and the structure of the energy spectrum work effectively as a wavelet signature. b. Two Gaussian eddies without noise From now on we will leave out noise since we have a basic understanding of it in the wavelet basis and since the energy levels of noise are extremely small.

7 MARCH 2002 LUO AND JAMESON 387 Table 2. Location of centers and averaging regions for three eddies at time 2. Eddy label A B C Eddy center 36.75N, E 34.75N, E 32.25N, E Neighborhood of eddy for averaging N, E N, E N, E d. The wavelet signature of a front Our final feature of interest is that of a front or discontinuity in the dataset. Locating discontinuities represents no challenge at all for wavelets and we will not spend much time on it. In a word, wavelets detect information at various scales and locations. Discontinuities have large amounts of energy at all scales and at one location and is straightforward to identify. In fact, if the front is a true discontinuity then the ump in the wavelet coefficients will be seen in all scales. Intuitively one would expect this since one way of viewing wavelet analysis is as a sequence of high-pass filters and a discontinuity will certainly have high-frequency energy at all scales. FIG. 5. The contour diagram of four eddies of different strength and size. The eddies are labeled A through D from the bottom to the top of the domain. The second entry in our catalog is a domain that contains two eddies, one significantly larger than the others, see Fig. 3 for the contour diagrams. In Fig. 4, we can see the wavelet energy spectrum for each of these two eddies. The top plot shows the energy profile for the smaller eddy and the bottom plot shows the energy profile for the larger eddy. One can see that these two eddies have significantly different energy profiles especially the energy levels at each wavelet scale. c. Four Gaussian eddies without noise The final entry in our catalog is an example with four eddies in the domain (see Fig. 5). Note that in this case the salient point is the ability of the technique to automatically find and separate these eddies. This is a key feature of the technique and the software implementation. In Fig. 6 we see the wavelet energy profiles for the four eddies. The energy levels at each scale for each eddy are quite distinct and there is no danger of ambiguity. 5. Identify, label, and track eddies in TOPEX/ Poseidon data This section is the culmination of the paper. Here we apply the previous developed method to real satellite data and illustrate that we can actually identify, label, and track real eddies. a. Eddies in a wavelet basis First of all, notice a simple wavelet decomposition of one snapshot of T/P data. In Fig. 7 we have plotted simply the magnitude of the wavelet coefficients. One can see that in the wavelet basis the eddies become easily observed. Our next challenge will be to identify, label, and track such eddies. b. Tracking real eddies The software implementation of this method easily finds and counts eddies in the domain. Here we choose three eddies from a full set of eddies and show that we can track the eddies and give a history of the evolution of their energy at various wavelet scales. We label our Table 1. Location of centers and averaging regions for three eddies at time 1. Table 3. Location of centers and averaging regions for three eddies at time 3. Eddy label Eddy center Neighborhood of eddy for averaging Eddy label Eddy center Neighborhood of eddy for averaging A B C 36.75N, E 35.25N, E 31.75N, E N, E N, E N, E A B C 36.75N, E 34.75N, E 32.25N, E N, E N, E N, E

8 388 JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY VOLUME 19 FIG. 6. The wavelet energy profiles of four eddies of different strength and size. three eddies A, B, and C; they have been selected in the first snapshot for tracking (see Fig. 8). The second and third snapshots illustrate that we can maintain a lock on the proper eddies without our scheme umping to other eddies. In Fig. 9 we give the wavelet energy profiles of the three eddies in the three different snapshots. The energy profiles are the amounts of energy at the wavelet scales from 1 to 6. For such energies to be found, we must define boxes around each of the eddies that contain most of the eddy energy. The bounds for these boxes are given in Tables 1 3. FIG. 7. The magnitude of D4 wavelet coefficients after one wavelet decomposition. One can see that the D4 wavelet clearly identifies the eddies and eddy centers. c. A dense field of eddies In a real application we would strive to analyze data from T/P that are essentially dense with eddies. That is,

9 MARCH 2002 LUO AND JAMESON 389 FIG. 8. Sea surface height contours showing eddies in three snapshots of T/P data at 10-day intervals. Three eddies have been identified, labeled (A, B, and C), and tracked. FIG. 9. The wavelet energy profiles of the three eddies labeled A (solid line), B (dash line), and C (dotted line) at three different times at 10-day intervals. in the entire two-dimensional ocean surface we would find ocean eddies at all locations and our technique would need to be robust in order to be able to handle such a challenging environment. In Fig. 10 we analyze such a dense field of eddies. We label and order the eddies from weakest to strongest, see Table 4 and we give the coordinates of a bounding box. Note that such a complex eddy field presents very little for the algorithm. One should note that if the eddy is near to the boundary of the domain, we elected not to attempt to label and measure the wavelet energy of such an eddy since it would be difficult to get a clean measure of the eddy s energy without corruption from the boundary. 6. Conclusions In this manuscript we have shown how wavelet analysis can be used to identify various structures that might FIG. 10. Fifteen eddies identified and labeled according to wavelet energy.

10 390 JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY VOLUME 19 Table 4. Fifteen eddies ranked according to wavelet energy at the finest scale. Columns 2 and 3 give the location of the eddy in plot 10. Eddy label Eddy center 37.25N, E 31.75N, E 34.75N, E 35.25N, E 36.25N, E 37.75N, E 36.25N, E 32.75N, E 37.25N, E 31.75N, E 34.75N, E 34.25N, E 34.75N, E 34.75N, E 36.75N, E Neighborhood of eddy for averaging N, E N, E N, E N, E N, E N, E N, E N, E N, E N, E N, E N, E N, E N, E N, E occur in a dataset. Such a dataset might come from observational data or perhaps from numerical simulations. We have tried to present a general approach to structure identification using wavelet analysis, but in the end we focused on one structure, the eddy, that seems to occur in many fields. Understanding and tracking eddies can provide valuable information about various physical processes especially mixing. We consider our technique to be nothing more than a first attempt at eddy tracking and in the future we expect to refine our method and to test it on more challenging datasets. The key advantage of the approach presented here is that it can easily be automated so that very large datasets can be analyzed without human input. Further, the scale information provided by wavelet analysis offers a very straightforward manner for a wavelet signature, which provides essentially a unique identifying mark to help to keep eddies separated. Also, the time evolution of the wavelet energy provides information on the transfer of turbulent energy in the eddies across scales. In short, the information provided by wavelet analysis cannot easily be obtained from other methods such as Fourier analysis or direct observation of contour plots. Acknowledgments. The two authors would like to express their appreciation to Professor Toshio Yamagata for his support and for bringing the two of us together for a successful collaboration. This work was performed under the auspices of the U.S. Department of Energy by the University of California Lawrence Livermore National Laboratory under Contract W-7405-Eng-48. REFERENCES Chui, C., 1992: Wavelets: A Tutorial in Theory and Applications. Vol. 2. Academic Press, Daubechies, I., 1992: Ten Lectures on Wavelets. Society for Industrial and Applied Mathematics, 357 pp., 1998: Orthonormal basis of compactly supported wavelets. Commun. Pure Appl. Math., 41, Erlebacher, G., M. Y. Hussaini, and L. Jameson, 1996: Wavelets: Theory and Applications. Oxford, 510 pp. Jameson, L., and T. Miyama, 2000: Wavelet analysis and ocean modeling: A dynamically adaptive numerical method WOFD- AHO. Mon. Wea. Rev., 128, Kirby, M., 2001: Geometic Data Analysis: An Empirical Approach to Dimensionality Reduction and the Study of Patterns. John Wiley and Sons. Meyer, Y., 1990: Ondelettes et Operators. Hermann, 215 pp. Saito, N., 1998: Least statistically-dependent basis and its application to image modeling. Wavelet Applications in Signal and Image Processing VI, A. F. Laine, M. A. Unser, and A. Aldroubi, Eds., The International Society for Optical Engineering, Strang, G., and T. Nguyen, 1996: Wavelets and Filter Banks. Wellesley-Cambridge Press, 490 pp.

Lectures notes. Rheology and Fluid Dynamics

Lectures notes. Rheology and Fluid Dynamics ÉC O L E P O L Y T E C H N IQ U E FÉ DÉR A L E D E L A U S A N N E Christophe Ancey Laboratoire hydraulique environnementale (LHE) École Polytechnique Fédérale de Lausanne Écublens CH-05 Lausanne Lectures

More information

An Introduction to Wavelets and some Applications

An Introduction to Wavelets and some Applications An Introduction to Wavelets and some Applications Milan, May 2003 Anestis Antoniadis Laboratoire IMAG-LMC University Joseph Fourier Grenoble, France An Introduction to Wavelets and some Applications p.1/54

More information

A Tutorial on Wavelets and their Applications. Martin J. Mohlenkamp

A Tutorial on Wavelets and their Applications. Martin J. Mohlenkamp A Tutorial on Wavelets and their Applications Martin J. Mohlenkamp University of Colorado at Boulder Department of Applied Mathematics mjm@colorado.edu This tutorial is designed for people with little

More information

Quadrature Prefilters for the Discrete Wavelet Transform. Bruce R. Johnson. James L. Kinsey. Abstract

Quadrature Prefilters for the Discrete Wavelet Transform. Bruce R. Johnson. James L. Kinsey. Abstract Quadrature Prefilters for the Discrete Wavelet Transform Bruce R. Johnson James L. Kinsey Abstract Discrepancies between the Discrete Wavelet Transform and the coefficients of the Wavelet Series are known

More information

Lecture 7 Multiresolution Analysis

Lecture 7 Multiresolution Analysis David Walnut Department of Mathematical Sciences George Mason University Fairfax, VA USA Chapman Lectures, Chapman University, Orange, CA Outline Definition of MRA in one dimension Finding the wavelet

More information

Lecture Notes 5: Multiresolution Analysis

Lecture Notes 5: Multiresolution Analysis Optimization-based data analysis Fall 2017 Lecture Notes 5: Multiresolution Analysis 1 Frames A frame is a generalization of an orthonormal basis. The inner products between the vectors in a frame and

More information

Quantum Mechanics-I Prof. Dr. S. Lakshmi Bala Department of Physics Indian Institute of Technology, Madras. Lecture - 21 Square-Integrable Functions

Quantum Mechanics-I Prof. Dr. S. Lakshmi Bala Department of Physics Indian Institute of Technology, Madras. Lecture - 21 Square-Integrable Functions Quantum Mechanics-I Prof. Dr. S. Lakshmi Bala Department of Physics Indian Institute of Technology, Madras Lecture - 21 Square-Integrable Functions (Refer Slide Time: 00:06) (Refer Slide Time: 00:14) We

More information

A Wavelet Optimized Adaptive Multi-Domain Method

A Wavelet Optimized Adaptive Multi-Domain Method NASA/CR-21747 ICASE Report No. 97-52 A Wavelet Optimized Adaptive Multi-Domain Method J. S. Hesthaven Brown University and L. M. Jameson ICASE Institute for Computer Applications in Science and Engineering

More information

MLISP: Machine Learning in Signal Processing Spring Lecture 8-9 May 4-7

MLISP: Machine Learning in Signal Processing Spring Lecture 8-9 May 4-7 MLISP: Machine Learning in Signal Processing Spring 2018 Prof. Veniamin Morgenshtern Lecture 8-9 May 4-7 Scribe: Mohamed Solomon Agenda 1. Wavelets: beyond smoothness 2. A problem with Fourier transform

More information

Wavelet Bases of the Interval: A New Approach

Wavelet Bases of the Interval: A New Approach Int. Journal of Math. Analysis, Vol. 1, 2007, no. 21, 1019-1030 Wavelet Bases of the Interval: A New Approach Khaled Melkemi Department of Mathematics University of Biskra, Algeria kmelkemi@yahoo.fr Zouhir

More information

Discrete Simulation of Power Law Noise

Discrete Simulation of Power Law Noise Discrete Simulation of Power Law Noise Neil Ashby 1,2 1 University of Colorado, Boulder, CO 80309-0390 USA 2 National Institute of Standards and Technology, Boulder, CO 80305 USA ashby@boulder.nist.gov

More information

Let p 2 ( t), (2 t k), we have the scaling relation,

Let p 2 ( t), (2 t k), we have the scaling relation, Multiresolution Analysis and Daubechies N Wavelet We have discussed decomposing a signal into its Haar wavelet components of varying frequencies. The Haar wavelet scheme relied on two functions: the Haar

More information

4.1 Haar Wavelets. Haar Wavelet. The Haar Scaling Function

4.1 Haar Wavelets. Haar Wavelet. The Haar Scaling Function 4 Haar Wavelets Wavelets were first aplied in geophysics to analyze data from seismic surveys, which are used in oil and mineral exploration to get pictures of layering in subsurface roc In fact, geophysicssts

More information

Designing Information Devices and Systems I Fall 2018 Lecture Notes Note Introduction to Linear Algebra the EECS Way

Designing Information Devices and Systems I Fall 2018 Lecture Notes Note Introduction to Linear Algebra the EECS Way EECS 16A Designing Information Devices and Systems I Fall 018 Lecture Notes Note 1 1.1 Introduction to Linear Algebra the EECS Way In this note, we will teach the basics of linear algebra and relate it

More information

Designing Information Devices and Systems I Spring 2018 Lecture Notes Note Introduction to Linear Algebra the EECS Way

Designing Information Devices and Systems I Spring 2018 Lecture Notes Note Introduction to Linear Algebra the EECS Way EECS 16A Designing Information Devices and Systems I Spring 018 Lecture Notes Note 1 1.1 Introduction to Linear Algebra the EECS Way In this note, we will teach the basics of linear algebra and relate

More information

Digital Image Processing

Digital Image Processing Digital Image Processing, 2nd ed. Digital Image Processing Chapter 7 Wavelets and Multiresolution Processing Dr. Kai Shuang Department of Electronic Engineering China University of Petroleum shuangkai@cup.edu.cn

More information

Module 4 MULTI- RESOLUTION ANALYSIS. Version 2 ECE IIT, Kharagpur

Module 4 MULTI- RESOLUTION ANALYSIS. Version 2 ECE IIT, Kharagpur Module 4 MULTI- RESOLUTION ANALYSIS Lesson Theory of Wavelets Instructional Objectives At the end of this lesson, the students should be able to:. Explain the space-frequency localization problem in sinusoidal

More information

Wavelet-Based Numerical Homogenization for Scaled Solutions of Linear Matrix Equations

Wavelet-Based Numerical Homogenization for Scaled Solutions of Linear Matrix Equations International Journal of Discrete Mathematics 2017; 2(1: 10-16 http://www.sciencepublishinggroup.com/j/dmath doi: 10.11648/j.dmath.20170201.13 Wavelet-Based Numerical Homogenization for Scaled Solutions

More information

Notes on Fourier Series and Integrals Fourier Series

Notes on Fourier Series and Integrals Fourier Series Notes on Fourier Series and Integrals Fourier Series et f(x) be a piecewise linear function on [, ] (This means that f(x) may possess a finite number of finite discontinuities on the interval). Then f(x)

More information

Wavelets and multiresolution representations. Time meets frequency

Wavelets and multiresolution representations. Time meets frequency Wavelets and multiresolution representations Time meets frequency Time-Frequency resolution Depends on the time-frequency spread of the wavelet atoms Assuming that ψ is centred in t=0 Signal domain + t

More information

Introduction to Discrete-Time Wavelet Transform

Introduction to Discrete-Time Wavelet Transform Introduction to Discrete-Time Wavelet Transform Selin Aviyente Department of Electrical and Computer Engineering Michigan State University February 9, 2010 Definition of a Wavelet A wave is usually defined

More information

Construction of Multivariate Compactly Supported Orthonormal Wavelets

Construction of Multivariate Compactly Supported Orthonormal Wavelets Construction of Multivariate Compactly Supported Orthonormal Wavelets Ming-Jun Lai Department of Mathematics The University of Georgia Athens, GA 30602 April 30, 2004 Dedicated to Professor Charles A.

More information

1 Introduction to Wavelet Analysis

1 Introduction to Wavelet Analysis Jim Lambers ENERGY 281 Spring Quarter 2007-08 Lecture 9 Notes 1 Introduction to Wavelet Analysis Wavelets were developed in the 80 s and 90 s as an alternative to Fourier analysis of signals. Some of the

More information

MATH Linear Algebra

MATH Linear Algebra MATH 304 - Linear Algebra In the previous note we learned an important algorithm to produce orthogonal sequences of vectors called the Gramm-Schmidt orthogonalization process. Gramm-Schmidt orthogonalization

More information

Piecewise constant approximation and the Haar Wavelet

Piecewise constant approximation and the Haar Wavelet Chapter Piecewise constant approximation and the Haar Wavelet (Group - Sandeep Mullur 4339 and Shanmuganathan Raman 433). Introduction Piecewise constant approximation principle forms the basis for the

More information

MODEL TYPE (Adapted from COMET online NWP modules) 1. Introduction

MODEL TYPE (Adapted from COMET online NWP modules) 1. Introduction MODEL TYPE (Adapted from COMET online NWP modules) 1. Introduction Grid point and spectral models are based on the same set of primitive equations. However, each type formulates and solves the equations

More information

SIO 211B, Rudnick, adapted from Davis 1

SIO 211B, Rudnick, adapted from Davis 1 SIO 211B, Rudnick, adapted from Davis 1 XVII.Empirical orthogonal functions Often in oceanography we collect large data sets that are time series at a group of locations. Moored current meter arrays do

More information

Chapter 7 Wavelets and Multiresolution Processing

Chapter 7 Wavelets and Multiresolution Processing Chapter 7 Wavelets and Multiresolution Processing Background Multiresolution Expansions Wavelet Transforms in One Dimension Wavelet Transforms in Two Dimensions Image Pyramids Subband Coding The Haar

More information

Wavelets and Multiresolution Processing. Thinh Nguyen

Wavelets and Multiresolution Processing. Thinh Nguyen Wavelets and Multiresolution Processing Thinh Nguyen Multiresolution Analysis (MRA) A scaling function is used to create a series of approximations of a function or image, each differing by a factor of

More information

Biorthogonal Spline Type Wavelets

Biorthogonal Spline Type Wavelets PERGAMON Computers and Mathematics with Applications 0 (00 1 0 www.elsevier.com/locate/camwa Biorthogonal Spline Type Wavelets Tian-Xiao He Department of Mathematics and Computer Science Illinois Wesleyan

More information

Wavelets and Signal Processing

Wavelets and Signal Processing Wavelets and Signal Processing John E. Gilbert Mathematics in Science Lecture April 30, 2002. Publicity Mathematics In Science* A LECTURE SERIES FOR UNDERGRADUATES Wavelets Professor John Gilbert Mathematics

More information

Quantum Mechanics - I Prof. Dr. S. Lakshmi Bala Department of Physics Indian Institute of Technology, Madras. Lecture - 9 Introducing Quantum Optics

Quantum Mechanics - I Prof. Dr. S. Lakshmi Bala Department of Physics Indian Institute of Technology, Madras. Lecture - 9 Introducing Quantum Optics Quantum Mechanics - I Prof. Dr. S. Lakshmi Bala Department of Physics Indian Institute of Technology, Madras Lecture - 9 Introducing Quantum Optics (Refer Slide Time: 00:07) In the last lecture I gave

More information

A Modular NMF Matching Algorithm for Radiation Spectra

A Modular NMF Matching Algorithm for Radiation Spectra A Modular NMF Matching Algorithm for Radiation Spectra Melissa L. Koudelka Sensor Exploitation Applications Sandia National Laboratories mlkoude@sandia.gov Daniel J. Dorsey Systems Technologies Sandia

More information

Digital Image Processing

Digital Image Processing Digital Image Processing Wavelets and Multiresolution Processing () Christophoros Nikou cnikou@cs.uoi.gr University of Ioannina - Department of Computer Science 2 Contents Image pyramids Subband coding

More information

Spanning Trees in Grid Graphs

Spanning Trees in Grid Graphs Spanning Trees in Grid Graphs Paul Raff arxiv:0809.2551v1 [math.co] 15 Sep 2008 July 25, 2008 Abstract A general method is obtained for finding recurrences involving the number of spanning trees of grid

More information

An Introduction to Wavelets

An Introduction to Wavelets 1 An Introduction to Wavelets Advanced Linear Algebra (Linear Algebra II) Heng-Yu Lin May 27 2013 2 Abstract With the prosperity of the Digital Age, information is nowadays increasingly, if not exclusively,

More information

Decomposition of Riesz frames and wavelets into a finite union of linearly independent sets

Decomposition of Riesz frames and wavelets into a finite union of linearly independent sets Decomposition of Riesz frames and wavelets into a finite union of linearly independent sets Ole Christensen, Alexander M. Lindner Abstract We characterize Riesz frames and prove that every Riesz frame

More information

Linear Algebra. Min Yan

Linear Algebra. Min Yan Linear Algebra Min Yan January 2, 2018 2 Contents 1 Vector Space 7 1.1 Definition................................. 7 1.1.1 Axioms of Vector Space..................... 7 1.1.2 Consequence of Axiom......................

More information

We have to prove now that (3.38) defines an orthonormal wavelet. It belongs to W 0 by Lemma and (3.55) with j = 1. We can write any f W 1 as

We have to prove now that (3.38) defines an orthonormal wavelet. It belongs to W 0 by Lemma and (3.55) with j = 1. We can write any f W 1 as 88 CHAPTER 3. WAVELETS AND APPLICATIONS We have to prove now that (3.38) defines an orthonormal wavelet. It belongs to W 0 by Lemma 3..7 and (3.55) with j =. We can write any f W as (3.58) f(ξ) = p(2ξ)ν(2ξ)

More information

7.5 Partial Fractions and Integration

7.5 Partial Fractions and Integration 650 CHPTER 7. DVNCED INTEGRTION TECHNIQUES 7.5 Partial Fractions and Integration In this section we are interested in techniques for computing integrals of the form P(x) dx, (7.49) Q(x) where P(x) and

More information

axioms Construction of Multiwavelets on an Interval Axioms 2013, 2, ; doi: /axioms ISSN

axioms Construction of Multiwavelets on an Interval Axioms 2013, 2, ; doi: /axioms ISSN Axioms 2013, 2, 122-141; doi:10.3390/axioms2020122 Article OPEN ACCESS axioms ISSN 2075-1680 www.mdpi.com/journal/axioms Construction of Multiwavelets on an Interval Ahmet Altürk 1 and Fritz Keinert 2,

More information

NOTES AND CORRESPONDENCE. A Quantitative Estimate of the Effect of Aliasing in Climatological Time Series

NOTES AND CORRESPONDENCE. A Quantitative Estimate of the Effect of Aliasing in Climatological Time Series 3987 NOTES AND CORRESPONDENCE A Quantitative Estimate of the Effect of Aliasing in Climatological Time Series ROLAND A. MADDEN National Center for Atmospheric Research,* Boulder, Colorado RICHARD H. JONES

More information

Page 52. Lecture 3: Inner Product Spaces Dual Spaces, Dirac Notation, and Adjoints Date Revised: 2008/10/03 Date Given: 2008/10/03

Page 52. Lecture 3: Inner Product Spaces Dual Spaces, Dirac Notation, and Adjoints Date Revised: 2008/10/03 Date Given: 2008/10/03 Page 5 Lecture : Inner Product Spaces Dual Spaces, Dirac Notation, and Adjoints Date Revised: 008/10/0 Date Given: 008/10/0 Inner Product Spaces: Definitions Section. Mathematical Preliminaries: Inner

More information

On the fast algorithm for multiplication of functions in the wavelet bases

On the fast algorithm for multiplication of functions in the wavelet bases Published in Proceedings of the International Conference Wavelets and Applications, Toulouse, 1992; Y. Meyer and S. Roques, edt., Editions Frontieres, 1993 On the fast algorithm for multiplication of functions

More information

Course and Wavelets and Filter Banks

Course and Wavelets and Filter Banks Course 8.327 and.30 Wavelets and Filter Banks Multiresolution Analysis (MRA): Requirements for MRA; Nested Spaces and Complementary Spaces; Scaling Functions and Wavelets Scaling Functions and Wavelets

More information

2 GOVERNING EQUATIONS

2 GOVERNING EQUATIONS 2 GOVERNING EQUATIONS 9 2 GOVERNING EQUATIONS For completeness we will take a brief moment to review the governing equations for a turbulent uid. We will present them both in physical space coordinates

More information

Hilbert Spaces. Hilbert space is a vector space with some extra structure. We start with formal (axiomatic) definition of a vector space.

Hilbert Spaces. Hilbert space is a vector space with some extra structure. We start with formal (axiomatic) definition of a vector space. Hilbert Spaces Hilbert space is a vector space with some extra structure. We start with formal (axiomatic) definition of a vector space. Vector Space. Vector space, ν, over the field of complex numbers,

More information

Non-stationary long memory parameter estimate based on wavelet

Non-stationary long memory parameter estimate based on wavelet Available online www.jocpr.com Journal of Chemical and Pharmaceutical Research, 2014, 6(7):2650-2654 Research Article ISSN : 0975-7384 CODEN(USA) : JCPRC5 Non-stationary long memory parameter estimate

More information

Spanning and Independence Properties of Finite Frames

Spanning and Independence Properties of Finite Frames Chapter 1 Spanning and Independence Properties of Finite Frames Peter G. Casazza and Darrin Speegle Abstract The fundamental notion of frame theory is redundancy. It is this property which makes frames

More information

Chapter 7 Wavelets and Multiresolution Processing. Subband coding Quadrature mirror filtering Pyramid image processing

Chapter 7 Wavelets and Multiresolution Processing. Subband coding Quadrature mirror filtering Pyramid image processing Chapter 7 Wavelets and Multiresolution Processing Wavelet transform vs Fourier transform Basis functions are small waves called wavelet with different frequency and limited duration Multiresolution theory:

More information

446 SCIENCE IN CHINA (Series F) Vol. 46 introduced in refs. [6, ]. Based on this inequality, we add normalization condition, symmetric conditions and

446 SCIENCE IN CHINA (Series F) Vol. 46 introduced in refs. [6, ]. Based on this inequality, we add normalization condition, symmetric conditions and Vol. 46 No. 6 SCIENCE IN CHINA (Series F) December 003 Construction for a class of smooth wavelet tight frames PENG Lizhong (Λ Π) & WANG Haihui (Ξ ) LMAM, School of Mathematical Sciences, Peking University,

More information

Wavelets and Image Compression. Bradley J. Lucier

Wavelets and Image Compression. Bradley J. Lucier Wavelets and Image Compression Bradley J. Lucier Abstract. In this paper we present certain results about the compression of images using wavelets. We concentrate on the simplest case of the Haar decomposition

More information

APPENDIX A. Background Mathematics. A.1 Linear Algebra. Vector algebra. Let x denote the n-dimensional column vector with components x 1 x 2.

APPENDIX A. Background Mathematics. A.1 Linear Algebra. Vector algebra. Let x denote the n-dimensional column vector with components x 1 x 2. APPENDIX A Background Mathematics A. Linear Algebra A.. Vector algebra Let x denote the n-dimensional column vector with components 0 x x 2 B C @. A x n Definition 6 (scalar product). The scalar product

More information

Vector Spaces. Vector space, ν, over the field of complex numbers, C, is a set of elements a, b,..., satisfying the following axioms.

Vector Spaces. Vector space, ν, over the field of complex numbers, C, is a set of elements a, b,..., satisfying the following axioms. Vector Spaces Vector space, ν, over the field of complex numbers, C, is a set of elements a, b,..., satisfying the following axioms. For each two vectors a, b ν there exists a summation procedure: a +

More information

Fourier Series and Integrals

Fourier Series and Integrals Fourier Series and Integrals Fourier Series et f(x) beapiece-wiselinearfunctionon[, ] (Thismeansthatf(x) maypossessa finite number of finite discontinuities on the interval). Then f(x) canbeexpandedina

More information

2 Voltage Potential Due to an Arbitrary Charge Distribution

2 Voltage Potential Due to an Arbitrary Charge Distribution Solution to the Static Charge Distribution on a Thin Wire Using the Method of Moments James R Nagel Department of Electrical and Computer Engineering University of Utah, Salt Lake City, Utah April 2, 202

More information

Physics I : Oscillations and Waves Prof. S. Bharadwaj Department of Physics and Meteorology Indian Institute of Technology, Kharagpur

Physics I : Oscillations and Waves Prof. S. Bharadwaj Department of Physics and Meteorology Indian Institute of Technology, Kharagpur Physics I : Oscillations and Waves Prof. S. Bharadwaj Department of Physics and Meteorology Indian Institute of Technology, Kharagpur Lecture - 21 Diffraction-II Good morning. In the last class, we had

More information

REGULARITY AND CONSTRUCTION OF BOUNDARY MULTIWAVELETS

REGULARITY AND CONSTRUCTION OF BOUNDARY MULTIWAVELETS REGULARITY AND CONSTRUCTION OF BOUNDARY MULTIWAVELETS FRITZ KEINERT Abstract. The conventional way of constructing boundary functions for wavelets on a finite interval is to form linear combinations of

More information

Linear Algebra, Summer 2011, pt. 2

Linear Algebra, Summer 2011, pt. 2 Linear Algebra, Summer 2, pt. 2 June 8, 2 Contents Inverses. 2 Vector Spaces. 3 2. Examples of vector spaces..................... 3 2.2 The column space......................... 6 2.3 The null space...........................

More information

Fourier-like Transforms

Fourier-like Transforms L 2 (R) Solutions of Dilation Equations and Fourier-like Transforms David Malone December 6, 2000 Abstract We state a novel construction of the Fourier transform on L 2 (R) based on translation and dilation

More information

COMS 4721: Machine Learning for Data Science Lecture 19, 4/6/2017

COMS 4721: Machine Learning for Data Science Lecture 19, 4/6/2017 COMS 4721: Machine Learning for Data Science Lecture 19, 4/6/2017 Prof. John Paisley Department of Electrical Engineering & Data Science Institute Columbia University PRINCIPAL COMPONENT ANALYSIS DIMENSIONALITY

More information

Lecture 10: Solving the Time-Independent Schrödinger Equation. 1 Stationary States 1. 2 Solving for Energy Eigenstates 3

Lecture 10: Solving the Time-Independent Schrödinger Equation. 1 Stationary States 1. 2 Solving for Energy Eigenstates 3 Contents Lecture 1: Solving the Time-Independent Schrödinger Equation B. Zwiebach March 14, 16 1 Stationary States 1 Solving for Energy Eigenstates 3 3 Free particle on a circle. 6 1 Stationary States

More information

Week 4: Differentiation for Functions of Several Variables

Week 4: Differentiation for Functions of Several Variables Week 4: Differentiation for Functions of Several Variables Introduction A functions of several variables f : U R n R is a rule that assigns a real number to each point in U, a subset of R n, For the next

More information

Linear Regression and Its Applications

Linear Regression and Its Applications Linear Regression and Its Applications Predrag Radivojac October 13, 2014 Given a data set D = {(x i, y i )} n the objective is to learn the relationship between features and the target. We usually start

More information

Design of Orthonormal Wavelet Filter Banks Using the Remez Exchange Algorithm

Design of Orthonormal Wavelet Filter Banks Using the Remez Exchange Algorithm Electronics and Communications in Japan, Part 3, Vol. 81, No. 6, 1998 Translated from Denshi Joho Tsushin Gakkai Ronbunshi, Vol. J80-A, No. 9, September 1997, pp. 1396 1402 Design of Orthonormal Wavelet

More information

Interference Between Distinguishable States. Thomas Alexander Meyer

Interference Between Distinguishable States. Thomas Alexander Meyer Interference Between Distinguishable States Thomas Alexander Meyer Interference effects are known to have a dependence upon indistinguishability of path. For this reason, it is accepted that different

More information

Module 7:Data Representation Lecture 35: Wavelets. The Lecture Contains: Wavelets. Discrete Wavelet Transform (DWT) Haar wavelets: Example

Module 7:Data Representation Lecture 35: Wavelets. The Lecture Contains: Wavelets. Discrete Wavelet Transform (DWT) Haar wavelets: Example The Lecture Contains: Wavelets Discrete Wavelet Transform (DWT) Haar wavelets: Example Haar wavelets: Theory Matrix form Haar wavelet matrices Dimensionality reduction using Haar wavelets file:///c /Documents%20and%20Settings/iitkrana1/My%20Documents/Google%20Talk%20Received%20Files/ist_data/lecture35/35_1.htm[6/14/2012

More information

On Sampling Errors in Empirical Orthogonal Functions

On Sampling Errors in Empirical Orthogonal Functions 3704 J O U R N A L O F C L I M A T E VOLUME 18 On Sampling Errors in Empirical Orthogonal Functions ROBERTA QUADRELLI, CHRISTOPHER S. BRETHERTON, AND JOHN M. WALLACE University of Washington, Seattle,

More information

Physics 221A Fall 1996 Notes 14 Coupling of Angular Momenta

Physics 221A Fall 1996 Notes 14 Coupling of Angular Momenta Physics 1A Fall 1996 Notes 14 Coupling of Angular Momenta In these notes we will discuss the problem of the coupling or addition of angular momenta. It is assumed that you have all had experience with

More information

Communication Engineering Prof. Surendra Prasad Department of Electrical Engineering Indian Institute of Technology, Delhi

Communication Engineering Prof. Surendra Prasad Department of Electrical Engineering Indian Institute of Technology, Delhi Communication Engineering Prof. Surendra Prasad Department of Electrical Engineering Indian Institute of Technology, Delhi Lecture - 41 Pulse Code Modulation (PCM) So, if you remember we have been talking

More information

LINEAR ALGEBRA KNOWLEDGE SURVEY

LINEAR ALGEBRA KNOWLEDGE SURVEY LINEAR ALGEBRA KNOWLEDGE SURVEY Instructions: This is a Knowledge Survey. For this assignment, I am only interested in your level of confidence about your ability to do the tasks on the following pages.

More information

A CLASS OF M-DILATION SCALING FUNCTIONS WITH REGULARITY GROWING PROPORTIONALLY TO FILTER SUPPORT WIDTH

A CLASS OF M-DILATION SCALING FUNCTIONS WITH REGULARITY GROWING PROPORTIONALLY TO FILTER SUPPORT WIDTH PROCEEDINGS OF THE AMERICAN MATHEMATICAL SOCIETY Volume 26, Number 2, December 998, Pages 350 3506 S 0002-9939(98)05070-9 A CLASS OF M-DILATION SCALING FUNCTIONS WITH REGULARITY GROWING PROPORTIONALLY

More information

Mathematics Department Stanford University Math 61CM/DM Inner products

Mathematics Department Stanford University Math 61CM/DM Inner products Mathematics Department Stanford University Math 61CM/DM Inner products Recall the definition of an inner product space; see Appendix A.8 of the textbook. Definition 1 An inner product space V is a vector

More information

MTH Linear Algebra. Study Guide. Dr. Tony Yee Department of Mathematics and Information Technology The Hong Kong Institute of Education

MTH Linear Algebra. Study Guide. Dr. Tony Yee Department of Mathematics and Information Technology The Hong Kong Institute of Education MTH 3 Linear Algebra Study Guide Dr. Tony Yee Department of Mathematics and Information Technology The Hong Kong Institute of Education June 3, ii Contents Table of Contents iii Matrix Algebra. Real Life

More information

CS168: The Modern Algorithmic Toolbox Lecture #11: The Fourier Transform and Convolution

CS168: The Modern Algorithmic Toolbox Lecture #11: The Fourier Transform and Convolution CS168: The Modern Algorithmic Toolbox Lecture #11: The Fourier Transform and Convolution Tim Roughgarden & Gregory Valiant May 8, 2015 1 Intro Thus far, we have seen a number of different approaches to

More information

MAT2342 : Introduction to Applied Linear Algebra Mike Newman, fall Projections. introduction

MAT2342 : Introduction to Applied Linear Algebra Mike Newman, fall Projections. introduction MAT4 : Introduction to Applied Linear Algebra Mike Newman fall 7 9. Projections introduction One reason to consider projections is to understand approximate solutions to linear systems. A common example

More information

Math Methods for Polymer Physics Lecture 1: Series Representations of Functions

Math Methods for Polymer Physics Lecture 1: Series Representations of Functions Math Methods for Polymer Physics ecture 1: Series Representations of Functions Series analysis is an essential tool in polymer physics and physical sciences, in general. Though other broadly speaking,

More information

Symmetric Wavelet Tight Frames with Two Generators

Symmetric Wavelet Tight Frames with Two Generators Symmetric Wavelet Tight Frames with Two Generators Ivan W. Selesnick Electrical and Computer Engineering Polytechnic University 6 Metrotech Center, Brooklyn, NY 11201, USA tel: 718 260-3416, fax: 718 260-3906

More information

CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE. From Exploratory Factor Analysis Ledyard R Tucker and Robert C. MacCallum

CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE. From Exploratory Factor Analysis Ledyard R Tucker and Robert C. MacCallum CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE From Exploratory Factor Analysis Ledyard R Tucker and Robert C. MacCallum 1997 65 CHAPTER 4 THE COMMON FACTOR MODEL IN THE SAMPLE 4.0. Introduction In Chapter

More information

Wavelets For Computer Graphics

Wavelets For Computer Graphics {f g} := f(x) g(x) dx A collection of linearly independent functions Ψ j spanning W j are called wavelets. i J(x) := 6 x +2 x + x + x Ψ j (x) := Ψ j (2 j x i) i =,..., 2 j Res. Avge. Detail Coef 4 [9 7

More information

Summary of free theory: one particle state: vacuum state is annihilated by all a s: then, one particle state has normalization:

Summary of free theory: one particle state: vacuum state is annihilated by all a s: then, one particle state has normalization: The LSZ reduction formula based on S-5 In order to describe scattering experiments we need to construct appropriate initial and final states and calculate scattering amplitude. Summary of free theory:

More information

An Adaptive Pseudo-Wavelet Approach for Solving Nonlinear Partial Differential Equations

An Adaptive Pseudo-Wavelet Approach for Solving Nonlinear Partial Differential Equations An Adaptive Pseudo-Wavelet Approach for Solving Nonlinear Partial Differential Equations Gregory Beylkin and James M. Keiser Wavelet Analysis and Applications, v.6, 1997, Academic Press. Contents 1 Introduction

More information

Vector analysis and vector identities by means of cartesian tensors

Vector analysis and vector identities by means of cartesian tensors Vector analysis and vector identities by means of cartesian tensors Kenneth H. Carpenter August 29, 2001 1 The cartesian tensor concept 1.1 Introduction The cartesian tensor approach to vector analysis

More information

Singular Value Decomposition. 1 Singular Value Decomposition and the Four Fundamental Subspaces

Singular Value Decomposition. 1 Singular Value Decomposition and the Four Fundamental Subspaces Singular Value Decomposition This handout is a review of some basic concepts in linear algebra For a detailed introduction, consult a linear algebra text Linear lgebra and its pplications by Gilbert Strang

More information

Wavelets in Scattering Calculations

Wavelets in Scattering Calculations Wavelets in Scattering Calculations W. P., Brian M. Kessler, Gerald L. Payne polyzou@uiowa.edu The University of Iowa Wavelets in Scattering Calculations p.1/43 What are Wavelets? Orthonormal basis functions.

More information

Select/Special Topics in Atomic Physics Prof. P. C. Deshmukh Department of Physics Indian Institute of Technology, Madras

Select/Special Topics in Atomic Physics Prof. P. C. Deshmukh Department of Physics Indian Institute of Technology, Madras Select/Special Topics in Atomic Physics Prof. P. C. Deshmukh Department of Physics Indian Institute of Technology, Madras Lecture - 9 Angular Momentum in Quantum Mechanics Dimensionality of the Direct-Product

More information

The New Graphic Description of the Haar Wavelet Transform

The New Graphic Description of the Haar Wavelet Transform he New Graphic Description of the Haar Wavelet ransform Piotr Porwik and Agnieszka Lisowska Institute of Informatics, Silesian University, ul.b dzi ska 39, 4-00 Sosnowiec, Poland porwik@us.edu.pl Institute

More information

Lecture Hilbert-Huang Transform. An examination of Fourier Analysis. Existing non-stationary data handling method

Lecture Hilbert-Huang Transform. An examination of Fourier Analysis. Existing non-stationary data handling method Lecture 12-13 Hilbert-Huang Transform Background: An examination of Fourier Analysis Existing non-stationary data handling method Instantaneous frequency Intrinsic mode functions(imf) Empirical mode decomposition(emd)

More information

Networks and Systems Prof V.G K. Murti Department of Electrical Engineering Indian Institute of Technology, Madras Lecture - 10 Fourier Series (10)

Networks and Systems Prof V.G K. Murti Department of Electrical Engineering Indian Institute of Technology, Madras Lecture - 10 Fourier Series (10) Networks and Systems Prof V.G K. Murti Department of Electrical Engineering Indian Institute of Technology, Madras Lecture - 10 Fourier Series (10) What we have seen in the previous lectures, is first

More information

16.4. Power Series. Introduction. Prerequisites. Learning Outcomes

16.4. Power Series. Introduction. Prerequisites. Learning Outcomes Power Series 6.4 Introduction In this Section we consider power series. These are examples of infinite series where each term contains a variable, x, raised to a positive integer power. We use the ratio

More information

Wavelets Marialuce Graziadei

Wavelets Marialuce Graziadei Wavelets Marialuce Graziadei 1. A brief summary 2. Vanishing moments 3. 2D-wavelets 4. Compression 5. De-noising 1 1. A brief summary φ(t): scaling function. For φ the 2-scale relation hold φ(t) = p k

More information

with a proper choice of the potential U(r). Clearly, we should include the potential of the ions, U ion (r):.

with a proper choice of the potential U(r). Clearly, we should include the potential of the ions, U ion (r):. The Hartree Equations So far we have ignored the effects of electron-electron (e-e) interactions by working in the independent electron approximation. In this lecture, we shall discuss how this effect

More information

Lecture 3: Haar MRA (Multi Resolution Analysis)

Lecture 3: Haar MRA (Multi Resolution Analysis) U U U WAVELETS AND MULTIRATE DIGITAL SIGNAL PROCESSING Lecture 3: Haar MRA (Multi Resolution Analysis) Prof.V.M.Gadre, EE, IIT Bombay 1 Introduction The underlying principle of wavelets is to capture incremental

More information

Math 261 Lecture Notes: Sections 6.1, 6.2, 6.3 and 6.4 Orthogonal Sets and Projections

Math 261 Lecture Notes: Sections 6.1, 6.2, 6.3 and 6.4 Orthogonal Sets and Projections Math 6 Lecture Notes: Sections 6., 6., 6. and 6. Orthogonal Sets and Projections We will not cover general inner product spaces. We will, however, focus on a particular inner product space the inner product

More information

Adapted Feature Extraction and Its Applications

Adapted Feature Extraction and Its Applications saito@math.ucdavis.edu 1 Adapted Feature Extraction and Its Applications Naoki Saito Department of Mathematics University of California Davis, CA 95616 email: saito@math.ucdavis.edu URL: http://www.math.ucdavis.edu/

More information

Multivector Calculus

Multivector Calculus In: J. Math. Anal. and Appl., ol. 24, No. 2, c Academic Press (1968) 313 325. Multivector Calculus David Hestenes INTRODUCTION The object of this paper is to show how differential and integral calculus

More information

Linear Algebra (part 1) : Vector Spaces (by Evan Dummit, 2017, v. 1.07) 1.1 The Formal Denition of a Vector Space

Linear Algebra (part 1) : Vector Spaces (by Evan Dummit, 2017, v. 1.07) 1.1 The Formal Denition of a Vector Space Linear Algebra (part 1) : Vector Spaces (by Evan Dummit, 2017, v. 1.07) Contents 1 Vector Spaces 1 1.1 The Formal Denition of a Vector Space.................................. 1 1.2 Subspaces...................................................

More information

Lecture Notes 1: Vector spaces

Lecture Notes 1: Vector spaces Optimization-based data analysis Fall 2017 Lecture Notes 1: Vector spaces In this chapter we review certain basic concepts of linear algebra, highlighting their application to signal processing. 1 Vector

More information

(Refer Slide Time: 0:18)

(Refer Slide Time: 0:18) Foundations of Wavelets, Filter Banks and Time Frequency Analysis. Professor Vikram M. Gadre. Department Of Electrical Engineering. Indian Institute of Technology Bombay. Week-1. Lecture -2.3 L2 Norm of

More information

Topics in Representation Theory: Fourier Analysis and the Peter Weyl Theorem

Topics in Representation Theory: Fourier Analysis and the Peter Weyl Theorem Topics in Representation Theory: Fourier Analysis and the Peter Weyl Theorem 1 Fourier Analysis, a review We ll begin with a short review of simple facts about Fourier analysis, before going on to interpret

More information