Applications of Wavelets and Framelets
|
|
- Vivien Joseph
- 5 years ago
- Views:
Transcription
1 Applications of Wavelets and Framelets Bin Han Department of Mathematical and Statistical Sciences University of Alberta, Edmonton, Canada Present at 2017 International Undergraduate Summer Enrichment Program at UofA July 19, 2017 Bin Han (University of Alberta) Applications of Wavelets UofA 1 / 47
2 Outline of Tutorial Wavelets in the function setting. Some applications of wavelets and framelets Tensor product wavelets and framelets Image processing using complex tight framelets. Subdivision schemes in computer graphics. Declaration: Some figures and graphs in this talk are from various sources from Internet, or from published papers, or produced by matlab, maple, or C programming. [Details and sources of all graphs can be provided upon request of the audience.] Bin Han (University of Alberta) Applications of Wavelets UofA 2 / 47
3 What Is a Wavelet in the Function Setting? Let φ = (φ 1,...,φ r ) T and ψ = (ψ 1,...,ψ s ) T in L 2 (R). A system is derived from φ,ψ via dilates and integer shifts: AS 0 (φ;ψ) :={φ( k) : k Z} {ψ j;k := 2 j/2 ψ(2 j k) : j N {0},k Z}. {φ;ψ} is called an orthogonal wavelet in L 2 (R) if AS 0 (φ;ψ) is an orthonormal basis of L 2 (R). {φ;ψ} is a tight framelet in L 2 (R) if f 2 L 2 (R) = k Z f,φ( k) 2 l 2 + f,ψ j;k 2 l 2, f L 2 (R). Orthogonal wavelet and tight framelet representation: f = f,ψ j;k ψ j;k, f L 2 (R), k Z f,φ( k) φ( k)+ j=0 j=0 k Z k Z where f,g := R f(x)g(x)t dx is the inner product. Bin Han (University of Alberta) Applications of Wavelets UofA 3 / 47
4 Dilates of a Wavelet 1.5 ψ 2;128 1 ψ 0; ψ 2; Bin Han (University of Alberta) Applications of Wavelets UofA 4 / 47
5 Integer Shifts of a Wavelet 0.5 ψ 0;0 ψ 0; ψ 2;0 ψ 2; Bin Han (University of Alberta) Applications of Wavelets UofA 5 / 47
6 Why Wavelets? A wavelet ψ often has 1 compact support good spatial localization. 2 high smoothness/regularity good frequency localization. 3 high vanishing moments multiscale sparse representation. 4 associated filter banks fast wavelet transform to compute coefficients f,ψ j;k through filter banks. 5 singularity detecting/locating and good approximation property. 6 close relations to windowed and fast Fourier transform. Explanation: Vanishing moments: x j,ψ(x) = 0 for j = 0,...,N. suppψ j;k = 2 j k +2 j suppψ 2 j k when j. f,ψ j;k = f P,ψ j;k 0 if f a polynomial P on suppψ j;k. If f,ψ j;k is large, then the singularity is around 2 j k. Bin Han (University of Alberta) Applications of Wavelets UofA 6 / 47
7 Tight Framelets or Orthogonal Wavelets Theorem: Let φ = (φ 1,...,φ r ) T and ψ = (ψ 1,...,ψ s ) T in L 2 (R). {φ;ψ} is a tight framelet (or orthogonal wavelet) in L 2 (R) 1 lim j φ(2 j ξ) 2 l 2 = 1; 2 there exist r r matrix â and s r matrix b of 2π-periodic measurable functions in L (T) such that φ(2ξ) = â(ξ) φ(ξ), i.e., φ = 2 a(k)φ(2 k), k Z ψ(2ξ) = b(ξ) φ(ξ), i.e., ψ = 2 k Zb(k)φ(2 k), and {â; b} is a tight framelet filter bank: [â(ξ) ] T â(ξ +π) â(ξ)t b(ξ) b(ξ) b(ξ +π) â(ξ +π) T T = I 2r, a.e.ξ R. b(ξ +π) 3 s = r and {φ( k)} k Z is an orthonormal system in L 2 (R), where f(ξ) := R f(x)e iξx dx and â(ξ) := k Z a(k)e ikξ. Bin Han (University of Alberta) Applications of Wavelets UofA 7 / 47
8 Example: Haar Orthogonal Wavelet {φ; ψ} Refinable function Wavelet φ = χ [0,1] ψ := χ [1/2,1] χ [0,1/2]. φ = φ(2 )+φ(2 1) ψ = φ(2 1) φ(2 ). φ and ψ have explicit expressions and φ is the B-spline of order 1. Bin Han (University of Alberta) Applications of Wavelets UofA 8 / 47
9 Example: Daubechies Orthogonal Wavelet {φ; ψ} φ = φ(2 ) φ(2 1) φ(2 2) φ(2 3). ψ = φ(2 ) φ(2 1) φ(2 2) φ(2 3). The functions φ and ψ do not have explicit expressions. Bin Han (University of Alberta) Applications of Wavelets UofA 9 / 47
10 Tensor Product (Separable) Tight Framelet Let {a;b 1,...,b s } be a 1D tight framelet filter bank. If s = 1, {a;b 1 } is called an orthonormal wavelet filter bank. Tensor product filters: [u 1 u d ](k 1,...,k d ) = u 1 (k 1 ) u d (k d ). Tensor product tight framelet filter bank: {a;b 1,...,b s } {a;b 1,...,b s }. Tensor product functions: [f 1 f d ](x 1,...,x d ) = f 1 (x 1 ) f d (x d ). Tensor product tight framelet: {φ;ψ 1,...,ψ s } {φ;ψ 1,...,ψ s }. Advantages: fast and simple algorithm. Bin Han (University of Alberta) Applications of Wavelets UofA 10 / 47
11 Tree Structure and Sparsity of Wavelet Coefficients Bin Han (University of Alberta) Applications of Wavelets UofA 11 / 47
12 Image Compression Using Orthogonal Wavelets Original Lena image and reconstructed Lena images with compression ratios 32 and 128 using SPIHT. Large coefficients are recorded with priority and tree structure is used. Bin Han (University of Alberta) Applications of Wavelets UofA 12 / 47
13 Image Denoising Using Orthogonal Wavelets Wavelet-shrinkage from statistics: small coefficients are set to 0. Bin Han (University of Alberta) Applications of Wavelets UofA 13 / 47
14 Curve Modeling: Corner Cutting Subdivision Scheme Initial control polygon v, iterated once S a v, iterated 5 times S 5 a v, where a = { 1 8, 3 8, 3 8, 1 8 } [0,3] is the B-spline filter of order 3. Bin Han (University of Alberta) Applications of Wavelets UofA 14 / 47
15 Surface Modeling by Subdivision scheme Initial mesh v, iterated once S a,m v, iterated twice S 2 a,m v. Bin Han (University of Alberta) Applications of Wavelets UofA 15 / 47
16 Subdivision Surfaces Used in Animated Movies Bin Han (University of Alberta) Applications of Wavelets UofA 16 / 47
17 Bandlimited Complex Tight Framelets TP-CTF π -π - π π 3π π 2 Tight framelet filter bank CTF 6 := {a +,a ;b 1 +,b+ 2,b 1,b 2 }: black lines for â+ and â ; dashed lines for b + 1 and b 1 ; dotted lines for b + 2 and b 2 : a := a +, b 1 := b+ 1, b 2 := b+ 2, and â + := χ [0,c];ε,ε, b+ 1 := χ [c 1,c 2 ];ε,ε, b+ 1 := χ [c 2,π];ε,ε. The tensor product tight framelet TP-CTF 6 := d CTF 6. Take advantages of wavelets and Discrete Cosine Transform. Bin Han (University of Alberta) Applications of Wavelets UofA 17 / 47
18 Two-dimensional TP-CTF 6 (14 directions) Bin Han (University of Alberta) Applications of Wavelets UofA 18 / 47
19 Denoising Comparison for Barbara Image DTCWT TP-CTF 6 UDWT TV Shearlet Redundancy N/A 49 σ = σ = σ = σ = σ = DTCWT=Dual Tree Complex Wavelet Transform. TP-CTF 6 =Han and Zhao, SIAM J. Imag. Sci. 7 (2014), UDWT=Undecimated Discrete Wavelet Transform. TV=Rudin-Osher-Fatemi (ROF) model using higher-order scheme. Shearlet=shearlet frames in W. Lim, IEEE T. Image Process., Measure of performance: PSNR = 10log MSE. The larger PSNR value the better performance. Bin Han (University of Alberta) Applications of Wavelets UofA 19 / 47
20 Remove Mixed Gaussian and Impulse Noises Gaussian and Pepper and Salt impulse noise. Cameraman: σ = 0, p = 0.3, PSNR = Lena: σ = 15, p = 0.5, PSNR = Gaussian and Random-valued impulse noises: Barbara: σ = 30, p = 0.2, PSNR = Peppers: σ = 20, p = 0.1, PSNR = Bin Han (University of Alberta) Applications of Wavelets UofA 20 / 47
21 Remove Gaussian & Pepper-and-Salt Noise AOP TP-CTF 6 AOP TP-CTF 6 AOP TP-CTF 6 σ p Cameraman House Peppers (1.88) (1.81) (0.82) (2.10) (1.85) (0.76) (1.80) (3.51) (1.43) (1.31) (2.68) (0.91) σ p Cameraman House Peppers (1.25) (5.13) (1.73) (1.59) (6.25) (2.02) (3.73) (4.49) (2.25) (2.74) (4.23) (1.72) AOP, TV-based, SIAM J. Imaging, 5 (2013), TP-CTF 6, Shen/Han/Braverman, J. Math. Imaging Vis., 54 (2016), Bin Han (University of Alberta) Applications of Wavelets UofA 21 / 47
22 Figure: Corrupted by text with σ = 20. Recovered with PSNR= Bin Han (University of Alberta) Applications of Wavelets UofA 22 / 47 Image Inpainting Using TP-CTF 6 Figure: 80% missing pixels. Recovered by our algorithm: PSNR=31.67.
23 Examples of Subdivision Curve Bin Han (University of Alberta) Applications of Wavelets UofA 23 / 47
24 Examples of Subdivision Curve Bin Han (University of Alberta) Applications of Wavelets UofA 24 / 47
25 Subdivision Schemes A dilation matrix M is a d d integer matrix such that all the eigenvalues of M are greater than one in modulus. Examples of dilation matrices: 2I d (dyadic), 3I d (ternary), M 2 = [ ], N 2 = [ ], M 3 = [ M 2 and N 2 are called the quincunx dilation matrices inducing the quincunx lattice M 2 Z2 = N 2 Z2 = {(j,k) Z 2 : j +k is even}. The subdivision operator S a,m : l(z d ) l(z d ) is [S a,m v](n) := det(m) k Z d v(k)a(n Mk), ]. where v = {v(k)} k Z d l(z d ). Bin Han (University of Alberta) Applications of Wavelets UofA 25 / 47
26 Subdivision Triplets: Symmetry is Necessary A symmetry group G is a finite set of d d integer matrices with determinants ±1 forming a group under matrix multiplication. A mask/filter a = {a(k)} k Z d : Z d R is G-symmetric with symmetry center c a if a(e(k c a )+c a ) = a(k), k Z d,e G. A dilation matrix M is compatible with G if MEM 1 G, E G. (a,m,g) is called a subdivision triplet if M is compatible with G and the mask a is G-symmetric. Bin Han (University of Alberta) Applications of Wavelets UofA 26 / 47
27 Subdivision Schemes Using Triplet (a, M, G) Subdivision scheme: calculate v n := Sa,M n v for n N and attach the value v n (k) at the point M n (k c a ), k Z d. The subdivision scheme converges if {v n } n=1 converges to a continuous function v for every bounded initial control mesh v. If the symmetry center c a = 0, it is called a primal subdivision scheme; otherwise, it is called a dual subdivision scheme. Proposition: For a subdivision triplet (a, M, G) with symmetry center c a, if â(0) = 1 with â(ξ) := k Z d a(k)e ik ξ, then φ(e( c φ )+c φ ) = φ E G with c φ := (M I d ) 1 c a, where φ is the M-refinable (or basis) function associated with the mask/filter a defined by φ(ξ) := j=1â((mt ) j ξ),ξ R d. Bin Han (University of Alberta) Applications of Wavelets UofA 27 / 47
28 Important Dilation Matrices Two important symmetry groups: { [ ] [ ] D 4 := ±,±,± { D 6 := ± [ ] 1 0,± 0 1 [ ] 0 1,± 1 1 [ ] 1 1,± 1 0 [ ] 0 1,± 1 0 [ ] 0 1,± 1 0 [ ]} 0 1, 1 0 [ ] 1 1,± 0 1 [ ]} D 4 for the quadrilateral mesh and D 6 for the triangular mesh. N is G-equivalent to M if N = EMF for some E,F G. N 2 is D 4-equivalent to M 2. Theorem: For a 2 2 real-valued matrix M, 1 if M is compatible with the symmetry group D 4, then M must be D 4 -equivalent to either ci 2 or cm 2 for some c R. 2 if M is compatible with the symmetry group D 6, then M must be D 6 -equivalent to either ci 2 or cm 3 for some c R. Bin Han (University of Alberta) Applications of Wavelets UofA 28 / 47
29 Quad and Triangular Meshes Figure: The quadrilateral mesh Z 2 Q (left) and the triangular mesh Z2 T (right). Bin Han (University of Alberta) Applications of Wavelets UofA 29 / 47
30 Definition of Linear-phase Moments Interpolation: [S a,m v](mk) = v(k) for all k Z and v l(z d ) a(0) = det(m) 1, a(mk) = 0, k Z d \{0}. Interpolation on Polynomials: [S a,m p](mk) = p(k M 1 c) for all k Z and all polynomials p with deg(p) < m a has linear-phase moments with phase c: â(ξ) = e ic ξ + O( ξ m ), ξ 0; Define lpm(a) = m with the highest possible m. a has order m sum rules: â(ξ +2πω) = O( ξ m ), ξ 0,ω Ω M \{0}, where Ω M := [0,1) d [(M T ) 1 Z d ]. Define sr(a,m) = m with the highest possible m. Note: If a has symmetry with symmetry center c a, then c = c a. Bin Han (University of Alberta) Applications of Wavelets UofA 30 / 47
31 Importance of Linear-phase Moments {a;b 1,...,b s } is called a tight M-framelet filter bank if â(ξ) 2 + b 1 (ξ) b s (ξ) 2 = 1, s â(ξ)â(ξ +2πω)+ b l (ξ) b l (ξ +2πω) = 0, ω Ω M \{0}. l=1 Called an orthogonal M-wavelet filter bank if s = det(m) 1. If det(m) = 2, then s = 1, Ω M = {0,ω}, and {a;b} is an orthogonal M-wavelet filter bank for some γ Z d \[MZ d ], â(ξ) 2 + â(ξ +2πω) 2 = 1, b(ξ) = e iγ ξâ(ξ +2πω). A filter b has n vanishing moments if b(ξ) = O( ξ n ) as ξ 0. We define vm(b) := n with the highest n. Theorem: If {a;b 1,...,b s } is a tight M-framelet filter bank and a has symmetry with symmetry center c a, then min(vm(b 1 ),...,vm(b s )) = min(sr(a), 1 2 lpm(a)). Bin Han (University of Alberta) Applications of Wavelets UofA 31 / 47
32 Tight Framelets and Wavelets A function ψ has n vanishing moments if ψ(ξ) = O( ξ n ) as ξ 0. We define vm(ψ) := n with the largest n. Theorem: If {a;b 1,...,b s } is a tight M-framelet filter bank with â(0) = 1, let φ(ξ) := j=1â((mt ) j ξ), ψ l (M T ξ) := b l (ξ) φ(ξ). Then {φ;ψ 1,...,ψ s } is a tight framelet in L 2 (R d ): f L 2 (R d ), f 2 L 2 (R d ) = f,φ( k) 2 + k Z d s f, det(m) j/2 ψ l (M j k) 2. j=0 l=1 k Z d vm(ψ l ) = vm(b l ) for all l = 1,...,s. It is a challenging problem to construct multivariate wavelets or tight framelets with symmetry and high vanishing moments. Bin Han (University of Alberta) Applications of Wavelets UofA 32 / 47
33 Fourier Transform For a function f on R d, its Fourier transform is defined to be f(ξ) := f(x)e iξ x dx, ξ R d. R d For a sequence a : Z d C, its Fourier series is â(ξ) := a(k)e ik ξ, ξ R d. k Z d Bin Han (University of Alberta) Applications of Wavelets UofA 33 / 47
34 Cascade Algorithms How to solve the refinement equation: φ = det(m) k Z d a(k)φ(m k), where the mask a : Z d R is finitely supported, equivalently, φ(ξ) = â((m T ) 1 ξ) φ((m T ) 1 ξ). Cascade algorithm: The cascade operator R is defined to be R a,m f := det(m) k Z d a(k)φ(m k). φ is a fixed point of R a,m by φ = R a,m φ. {f n := R n a,m f} n N of functions is called a cascade algorithm. The cascade algorithm converges if for every compactly supported eligible initial function f, there exists a continuous function f such that lim n f n f C(R d ) = 0. Bin Han (University of Alberta) Applications of Wavelets UofA 34 / 47
35 Cascade Algorithm and Subdivision Schemes Cascade algorithm: the iterative sequence {f n := R n a,m f} n N of functions. Subdivision scheme: calculate v n := Sa,M n v for n N and attach the value v n (k) at the point M n (k c a ), k Z d. Relation: f n = R n a,m f = k Z d [S n a,m δ](k)f(mn k), where δ is the Dirac sequence such that δ(0) = 1 and δ(k) = 0 for all k 0. Let h be the hat function (in 1d, h = max(1 x,0)). Then connecting points of v n be flat pieces to form a function g n is equivalent to (assume c a = 0) g n = R n a,mf with f := k Z d v(k)h( k). Bin Han (University of Alberta) Applications of Wavelets UofA 35 / 47
36 Role of a Dilation Matrix Figure: represents vertices in the coarse mesh Z 2 and represents new vertices in the refinement mesh M 1 Z 2. The M-refinement of the reference mesh Z 2, from left to right, are for subdivision triplets (a,2i 2,D 4 ), (a,m 2,D 4), (a,2i 2,D 6 ), and (a,m 3,D 6), where M 2 and M 3. Bin Han (University of Alberta) Applications of Wavelets UofA 36 / 47
37 Implemented by Convolution Subdivision scheme: calculate v n := Sa,M n v for n N and attach the value v n (k) at the point M n (k c a ), k Z d. For β,γ Z d, [S a,m v](γ +Mβ) = det(m) k Z d v(k)a(γ +Mβ Mk) = det(m) [v a [γ:m] ](β), where the coset mask a [γ:m] of the mask a is defined to be a [γ:m] (k) := a(γ +Mk), k,γ Z d. Local averaging: det(m) k Z d a [γ:m] (k) = 1 for all γ Z d. The value [S a,m v](γ +Mβ) = v(β+ ), det(m) a [γ:m] ( ), is put at β +M 1 γ M 1 c a. M 1 γ-stencil of the mask a: { det(m) a(γ Mk)} k Z d. Bin Han (University of Alberta) Applications of Wavelets UofA 37 / 47
38 1D Subdivision Triplets For a finitely supported sequence a : Z R, we define a(z) := k Z a(k)z k, z C\{0}. Let M be an integer greater than one. Subdivision operator: [S a,m v](z) = Mv(z 2 )a(z). a has order n sum rules if and only if a(z) = (1+z + +z M 1 ) n b(z) for some Laurent polynomial b. a has order n linear-phase moments if and only if a(z) = z c + O( z 1 n ), z 1. a is interpolatory with respect to M if a(0) = 1, a(mk) = 0, k Z\{0}. M Bin Han (University of Alberta) Applications of Wavelets UofA 38 / 47
39 1D Subdivision Triplet The triplet (a,2,{ 1,1}) is a primal subdivision triplet with where a = 1 2 {w 3,w 2,w 1,w 0,w 1,w 2,w 3 } [ 3,3], w 0 = 3+t 4, w 1 = 8+t 16, w 2 = 1 t 8, w 3 = t 16 with t R. If t = 1, then a = 2 ab 6 ( 3) and sr(a,2) = 6, lpm(a) = 2 and sm p (a,2) = 5+1/p for all 1 p. If t 1/2, then sr(a,2) = 4. sm (a,2) = 3 log 2 (1+t) provided t > 1/2. We only have sm (a,2) 3 log 2 t for t 1/2. When t = 0, a = a4( 2) B is the centered B-spline filter of order 4 with sr(a,2) = 4 and lpm(a) = 2. When t = 1, a is an interpolatory 2-wavelet filter with sr(a,2) = 4 and lpm(a) = 4. Bin Han (University of Alberta) Applications of Wavelets UofA 39 / 47
40 Subdivision Stencils w 2 w 0 w 2 w 3 w 1 w 1 w 3 Figure: The 0-stencil (left) and the 1 2-stencil (right) of the primal subdivision scheme. It is an interpolatory 2-wavelet filter if w 2 = 1 t 8 = 0 (i.e. t = 1). Since M = 2, each line segment (with endpoints ) in the coarse mesh Z is equally split into two line segments with one new vertex ( ) in the middle. Bin Han (University of Alberta) Applications of Wavelets UofA 40 / 47
41 1D Subdivision Triplet The triplet (a,2,{ 1,1}) is a dual subdivision triplet with where a = 1 2 {w 2,w 1,w 0,w 0,w 1,w 2 } [ 2,3], w 0 = 12+3t 16, w 1 = 8 3t 32, w 2 = 3t 32 with t R. If t = 2, then a = 3 ab 5 ( 2) and sr(a,2) = 5,lpm(a) = 2 and sm p (a,2) = 4+1/p for all 1 p. sr(a,2) = 3 and sm (a,2) = 4 log 2 (4+3t) provided t > 2/3. We only have sm (a,2) 1 log 2 (3 t ) for t 2/3. When t = 0, a = a3 B( 1) is the shifted B-spline filter of order 3 with sr(a,2) = 3 and lpm(a) = 2. When t = 1, sr(a,2) = 3 and lpm(a) = 4. Bin Han (University of Alberta) Applications of Wavelets UofA 41 / 47
42 Subdivision Stencils w 1 w 0 w 2 w 2 w 0 w 1 Figure: The 0-stencil (left) and the 1 2-stencil (right) of the dual subdivision scheme. The 1 2-stencil is the same as the 0-stencil. The value [S a,2 v](k) for k Z is attached to the center k 1 2 of the line segment [k 1,k] instead of the vertex k 2. Since M = 2, each line segment is equally split into two. Bin Han (University of Alberta) Applications of Wavelets UofA 42 / 47
43 1D Subdivision Triplet The triplet (a,3,{ 1,1}) is a primal subdivision triplet with where a = 1 3 {w 5,w 4,w 3,w 2,w 1,w 0,w 1,w 2,w 3,w 4,w 5 } [ 5,5], w 0 = 7 2t 1 8t 2, w 9 1 = 6 2t 1 5t 2, w 9 2 = 3+t 1+t 2, 9 w 3 = 1+t 1+4t 2, w 9 4 = t 1+3t 2 with t, w 9 5 = t 2 1,t 2 R. 9 If t 1 = 2/9 and t 2 = 1/9, then sr(a,3) = 5 and sm p (a,3) = 4+1/p for all 1 p whose 3-refinable function is the B-spline of order 5. sm (a,2) 2 log 3 max( 1 2t 1 2t 2, 2t 1, 2t 2 ). If t 1 = 7/9 and t 2 = 4/9, then a is an interpolatory 3-wavelet filter with sr(a,3) = 4 = lpm(a) and sm (a,3) log If t 1 = 5/11 and t 2 = 4/11, then a is an interpolatory 3-wavelet filter with sr(a,3) = 3 = lpm(a) and sm (a,3) 2+log 3 (11/10) > 2. Bin Han (University of Alberta) Applications of Wavelets UofA 43 / 47
44 1D Subdivision Triplet w 3 w 0 w 3 w 4 w 1 w 2 w 5 w 5 w 2 w 1 w 4 Figure: The 0-stencil (left), the 1 3 -stencil (middle), and 2 3-stencil of the subdivision scheme. Due to symmetry, 2 3-stencil is the same as the 1 3 -stencil. It is an interpolatory 3-wavelet filter if w 3 = 1+t 1+4t 2 9 = 0. Since M = 3, each line segment (with endpoints ) is equally split into three line segments with two new inserted vertices ( ) at 1 3 +Z and 2 3 +Z. Bin Han (University of Alberta) Applications of Wavelets UofA 44 / 47
45 Examples of Subdivision Curve Bin Han (University of Alberta) Applications of Wavelets UofA 45 / 47
46 Examples of Subdivision Curve Bin Han (University of Alberta) Applications of Wavelets UofA 46 / 47
47 Masks Used Subdivision curves at levels 1, 2, 3 with the initial control polygons at the first row. (1) uses the subdivision triplet (a,2,{ 1,1}) with a = a B 4 ( 2) (2) uses interpolatory subdivision triplet (a, 2,{ 1, 1}). (3) uses (a,2,{ 1,1}) with a = a B 3( 1). (4) the corner cutting scheme (5) uses (a,3,{ 1,1}). (6) uses interpolatory (a, 3,{ 1, 1}). Bin Han (University of Alberta) Applications of Wavelets UofA 47 / 47
Edge Detections Using Box Spline Tight Frames
Edge Detections Using Box Spline Tight Frames Ming-Jun Lai 1) Abstract. We present a piece of numerical evidence that box spline tight frames are very useful for edge detection of images. Comparsion with
More informationComputation of interpolatory splines via triadic subdivision
Computation of interpolatory splines via triadic subdivision Valery A. Zheludev and Amir Z. Averbuch Abstract We present an algorithm for computation of interpolatory splines of arbitrary order at triadic
More informationHighly Symmetric Bi-frames for Triangle Surface Multiresolution Processing
Highly Symmetric Bi-frames for Triangle Surface Multiresolution Processing Qingtang Jiang and Dale K. Pounds Abstract In this paper we investigate the construction of dyadic affine (wavelet) bi-frames
More informationRefinable bivariate quartic and quintic C 2 -splines for quadrilateral subdivisions
Refinable bivariate quartic and quintic C 2 -splines for quadrilateral subdivisions Charles K. Chui, Qingtang Jiang Department of Mathematics and Computer Science University of Missouri St. Louis St. Louis,
More informationMatrix-valued 4-point Spline and 3-point Non-spline Interpolatory Curve Subdivision Schemes
Matrix-valued 4-point Spline and -point Non-spline Interpolatory Curve Subdivision Schemes Charles K. Chui, Qingtang Jiang Department of Mathematics and Computer Science University of Missouri St. Louis
More informationA C 2 Four-Point Subdivision Scheme with Fourth Order Accuracy and its Extensions
A C 2 Four-Point Subdivision Scheme with Fourth Order Accuracy and its Extensions Nira Dyn School of Mathematical Sciences Tel Aviv University Michael S. Floater Department of Informatics University of
More informationA C 2 Four-Point Subdivision Scheme with Fourth Order Accuracy and its Extensions
A C 2 Four-Point Subdivision Scheme with Fourth Order Accuracy and its Extensions Nira Dyn Michael S. Floater Kai Hormann Abstract. We present a new four-point subdivision scheme that generates C 2 curves.
More informationA Wavelet Tour of Signal Processing The Sparse Way
A Wavelet Tour of Signal Processing The Sparse Way Stephane Mallat with contributions from Gabriel Peyre AMSTERDAM BOSTON HEIDELBERG LONDON NEWYORK OXFORD PARIS SAN DIEGO SAN FRANCISCO SINGAPORE SYDNEY»TOKYO
More informationErasing Haar Coefficients
Recap Haar simple and fast wavelet transform Limitations not smooth enough: blocky How to improve? classical approach: basis functions Lifting: transforms 1 Erasing Haar Coefficients 2 Page 1 Classical
More informationTERM PAPER ON The Compressive Sensing Based on Biorthogonal Wavelet Basis
TERM PAPER ON The Compressive Sensing Based on Biorthogonal Wavelet Basis Submitted By: Amrita Mishra 11104163 Manoj C 11104059 Under the Guidance of Dr. Sumana Gupta Professor Department of Electrical
More information3. Lifting Scheme of Wavelet Transform
3. Lifting Scheme of Wavelet Transform 3. Introduction The Wim Sweldens 76 developed the lifting scheme for the construction of biorthogonal wavelets. The main feature of the lifting scheme is that all
More informationCoE4TN3 Image Processing. Wavelet and Multiresolution Processing. Image Pyramids. Image pyramids. Introduction. Multiresolution.
CoE4TN3 Image Processing Image Pyramids Wavelet and Multiresolution Processing 4 Introduction Unlie Fourier transform, whose basis functions are sinusoids, wavelet transforms are based on small waves,
More informationSubdivision Curves and Surfaces
Subdivision Surfaces or How to Generate a Smooth Mesh?? Subdivision Curves and Surfaces Subdivision given polyline(2d)/mesh(3d) recursively modify & add vertices to achieve smooth curve/surface Each iteration
More informationContourlets: Construction and Properties
Contourlets: Construction and Properties Minh N. Do Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign www.ifp.uiuc.edu/ minhdo minhdo@uiuc.edu Joint work with
More informationBeyond Wavelets: Directional Multiresolution Image Representation
Beyond Wavelets: Directional Multiresolution Image Representation Minh N. Do Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign www.ifp.uiuc.edu/ minhdo minhdo@uiuc.edu
More informationComparative Analysis of Image Compression Using Wavelet and Ridgelet Transform
Comparative Analysis of Image Compression Using Wavelet and Ridgelet Transform Thaarini.P 1, Thiyagarajan.J 2 PG Student, Department of EEE, K.S.R College of Engineering, Thiruchengode, Tamil Nadu, India
More informationApplication of Daubechies Wavelets for Image Compression
Application of Daubechies Wavelets for Image Compression Heydari. Aghile 1,*, Naseri.Roghaye 2 1 Department of Math., Payame Noor University, Mashad, IRAN, Email Address a_heidari@pnu.ac.ir, Funded by
More informationRobust image recovery via total-variation minimization
Robust image recovery via total-variation minimization Rachel Ward University of Texas at Austin (Joint work with Deanna Needell, Claremont McKenna College) February 16, 2012 2 Images are compressible
More informationIMAGE ANALYSIS, CLASSIFICATION, and CHANGE DETECTION in REMOTE SENSING
SECOND EDITION IMAGE ANALYSIS, CLASSIFICATION, and CHANGE DETECTION in REMOTE SENSING ith Algorithms for ENVI/IDL Morton J. Canty с*' Q\ CRC Press Taylor &. Francis Group Boca Raton London New York CRC
More informationImage Transformation Techniques Dr. Rajeev Srivastava Dept. of Computer Engineering, ITBHU, Varanasi
Image Transformation Techniques Dr. Rajeev Srivastava Dept. of Computer Engineering, ITBHU, Varanasi 1. Introduction The choice of a particular transform in a given application depends on the amount of
More informationFourier Transform of Bernstein-Bézier Polynomials
Fourier Transform of Bernstein-Bézier Polynomials Charles. K. Chui, Tian-Xiao He 2, and Qingtang Jiang 3 Department of Mathematics and Computer Science University of Missouri-St. Louis, St. Louis, MO 632,
More informationTangents and curvatures of matrix-valued subdivision curves and their applications to curve design
Tangents and curvatures of matrix-valued subdivision curves and their applications to curve design Qingtang Jiang, James J. Smith Department of Mathematics and Computer Science University of Missouri St.
More informationGeneralized Tree-Based Wavelet Transform and Applications to Patch-Based Image Processing
Generalized Tree-Based Wavelet Transform and * Michael Elad The Computer Science Department The Technion Israel Institute of technology Haifa 32000, Israel *Joint work with A Seminar in the Hebrew University
More informationImage denoising using curvelet transform: an approach for edge preservation
Journal of Scientific & Industrial Research Vol. 3469, January 00, pp. 34-38 J SCI IN RES VOL 69 JANUARY 00 Image denoising using curvelet transform: an approach for edge preservation Anil A Patil * and
More informationCHAPTER 3 DIFFERENT DOMAINS OF WATERMARKING. domain. In spatial domain the watermark bits directly added to the pixels of the cover
38 CHAPTER 3 DIFFERENT DOMAINS OF WATERMARKING Digital image watermarking can be done in both spatial domain and transform domain. In spatial domain the watermark bits directly added to the pixels of the
More informationA Study on Subdivision Scheme-Draft. Kwan Pyo Ko Dongseo University Busan, South Korea
A Study on Subdivision Scheme-Draft Kwan Pyo Ko Dongseo University Busan, South Korea April 30, 007 Contents Introduction 9 Subdivision of Univariate Data 3. Definitions and Basic Results.............................
More informationFourier transform of Bernstein Bézier polynomials
Illinois Wesleyan University From the SelectedWorks of Tian-Xiao He 27 Fourier transform of Bernstein Bézier polynomials Tian-Xiao He, Illinois Wesleyan University Charles K. Chui Qingtang Jiang Available
More informationEdge and local feature detection - 2. Importance of edge detection in computer vision
Edge and local feature detection Gradient based edge detection Edge detection by function fitting Second derivative edge detectors Edge linking and the construction of the chain graph Edge and local feature
More informationOutline Introduction Problem Formulation Proposed Solution Applications Conclusion. Compressed Sensing. David L Donoho Presented by: Nitesh Shroff
Compressed Sensing David L Donoho Presented by: Nitesh Shroff University of Maryland Outline 1 Introduction Compressed Sensing 2 Problem Formulation Sparse Signal Problem Statement 3 Proposed Solution
More informationSubdivision of Curves and Surfaces: An Overview
Subdivision of Curves and Surfaces: An Overview Ben Herbst, Karin M Hunter, Emile Rossouw Applied Mathematics, Department of Mathematical Sciences, University of Stellenbosch, Private Bag X1, Matieland,
More informationRegularity Analysis of Non Uniform Data
Regularity Analysis of Non Uniform Data Christine Potier and Christine Vercken Abstract. A particular class of wavelet, derivatives of B-splines, leads to fast and ecient algorithms for contours detection
More informationTwo Algorithms for Adaptive Approximation of Bivariate Functions by Piecewise Linear Polynomials on Triangulations
Two Algorithms for Adaptive Approximation of Bivariate Functions by Piecewise Linear Polynomials on Triangulations Nira Dyn School of Mathematical Sciences Tel Aviv University, Israel First algorithm from
More informationContents. Implementing the QR factorization The algebraic eigenvalue problem. Applied Linear Algebra in Geoscience Using MATLAB
Applied Linear Algebra in Geoscience Using MATLAB Contents Getting Started Creating Arrays Mathematical Operations with Arrays Using Script Files and Managing Data Two-Dimensional Plots Programming in
More informationData-Driven Tight Frame for Multi-Channel Images and Its Application to Joint Color-Depth Image Reconstruction
Data-Driven Tight Frame for Multi-Channel Images and Its Application to Joint Color-Depth Image Reconstruction Jin Wang Jian-Feng Cai Abstract In image restoration, we usually assume that the underlying
More informationCHAPTER 3 WAVELET DECOMPOSITION USING HAAR WAVELET
69 CHAPTER 3 WAVELET DECOMPOSITION USING HAAR WAVELET 3.1 WAVELET Wavelet as a subject is highly interdisciplinary and it draws in crucial ways on ideas from the outside world. The working of wavelet in
More informationShift-invariance in the Discrete Wavelet Transform
Shift-invariance in the Discrete Wavelet Transform Andrew P. Bradley Cooperative Research Centre for Sensor Signal and Information Processing, School of Information Technology and Electrical Engineering,
More informationAdvanced Geometric Modeling CPSC789
Advanced Geometric Modeling CPSC789 Fall 2004 General information about the course CPSC 789 Advanced Geometric Modeling Fall 2004 Lecture Time and Place ENF 334 TR 9:30 10:45 Instructor : Office: MS 618
More informationBOX SPLINE WAVELET FRAMES FOR IMAGE EDGE ANALYSIS
BOX SPLINE WAVELET FRAMES FOR IMAGE EDGE ANALYSIS WEIHONG GUO AND MING-JUN LAI Abstract. We present a new box spline wavelet frame and apply it for image edge analysis. The wavelet frame is constructed
More informationAn interpolating 4-point C 2 ternary stationary subdivision scheme
Computer Aided Geometric Design 9 (2002) 8 www.elsevier.com/locate/comaid An interpolating 4-point C 2 ternary stationary subdivision scheme M.F Hassan a,, I.P. Ivrissimitzis a, N.A. Dodgson a,m.a.sabin
More informationINF3320 Computer Graphics and Discrete Geometry
INF3320 Computer Graphics and Discrete Geometry More smooth Curves and Surfaces Christopher Dyken, Michael Floater and Martin Reimers 10.11.2010 Page 1 More smooth Curves and Surfaces Akenine-Möller, Haines
More informationQuad/triangle Subdivision, Nonhomogeneous Refinement Equation and Polynomial Reproduction
Quad/triangle Subdivision, Nonhomogeneous Refinement Equation and Polynomial Reproduction Qingtang Jiang a,, Baobin Li b, a Department of Mathematics and Computer Science, University of Missouri St. Louis,
More informationEdge Detection by Multi-Dimensional Wavelets
Edge Detection by Multi-Dimensional Wavelets Marlana Anderson, Chris Brasfield, Pierre Gremaud, Demetrio Labate, Katherine Maschmeyer, Kevin McGoff, Julie Siloti August 1, 2007 Abstract It is well known
More informationIMAGE DENOISING USING FRAMELET TRANSFORM
IMAGE DENOISING USING FRAMELET TRANSFORM Ms. Jadhav P.B. 1, Dr.Sangale.S.M. 2 1,2, Electronics Department,Shivaji University, (India) ABSTRACT Images are created to record or display useful information
More informationELEC639B Term Project: An Image Compression System with Interpolating Filter Banks
1 ELEC639B Term Project: An Image Compression System with Interpolating Filter Banks Yi Chen Abstract In this project, two families of filter banks are constructed, then their performance is compared with
More informationCSEP 521 Applied Algorithms Spring Lossy Image Compression
CSEP 521 Applied Algorithms Spring 2005 Lossy Image Compression Lossy Image Compression Methods Scalar quantization (SQ). Vector quantization (VQ). DCT Compression JPEG Wavelet Compression SPIHT UWIC (University
More informationResearch Article A Family of Even-Point Ternary Approximating Schemes
International Scholarly Research Network ISRN Applied Mathematics Volume, Article ID 97, pages doi:.5//97 Research Article A Family of Even-Point Ternary Approximating Schemes Abdul Ghaffar and Ghulam
More informationDaubechies Wavelets and Interpolating Scaling Functions and Application on PDEs
Daubechies Wavelets and Interpolating Scaling Functions and Application on PDEs R. Schneider F. Krüger TUB - Technical University of Berlin November 22, 2007 R. Schneider F. Krüger Daubechies Wavelets
More informationAMS527: Numerical Analysis II
AMS527: Numerical Analysis II A Brief Overview of Finite Element Methods Xiangmin Jiao SUNY Stony Brook Xiangmin Jiao SUNY Stony Brook AMS527: Numerical Analysis II 1 / 25 Overview Basic concepts Mathematical
More informationContents. I The Basic Framework for Stationary Problems 1
page v Preface xiii I The Basic Framework for Stationary Problems 1 1 Some model PDEs 3 1.1 Laplace s equation; elliptic BVPs... 3 1.1.1 Physical experiments modeled by Laplace s equation... 5 1.2 Other
More informationTwo Algorithms for Approximation in Highly Complicated Planar Domains
Two Algorithms for Approximation in Highly Complicated Planar Domains Nira Dyn and Roman Kazinnik School of Mathematical Sciences, Tel-Aviv University, Tel-Aviv 69978, Israel, {niradyn,romank}@post.tau.ac.il
More informationYves Nievergelt. Wavelets Made Easy. Springer Science+Business Media, LLC
Wavelets Made Easy Yves Nievergelt Wavelets Made Easy Springer Science+Business Media, LLC Yves Nievergelt Department of Mathematics Eastem Washington University Cheney, WA 99004-2431 USA Library of Congress
More informationDirectional multiscale processing of images using wavelets with composite dilations
Directional multiscale processing of images using wavelets with composite dilations Glenn R. Easley System Planning Corporation, Arlington, VA Demetrio Labate Department of Mathematics, University of Houston,
More informationDiffusion Wavelets for Natural Image Analysis
Diffusion Wavelets for Natural Image Analysis Tyrus Berry December 16, 2011 Contents 1 Project Description 2 2 Introduction to Diffusion Wavelets 2 2.1 Diffusion Multiresolution............................
More informationSubdivision Surfaces
Subdivision Surfaces CS 4620 Lecture 31 Cornell CS4620 Fall 2015 1 Administration A5 due on Friday Dreamworks visiting Thu/Fri Rest of class Surfaces, Animation, Rendering w/ prior instructor Steve Marschner
More informationA NEW PROOF OF THE SMOOTHNESS OF 4-POINT DESLAURIERS-DUBUC SCHEME
J. Appl. Math. & Computing Vol. 18(2005), No. 1-2, pp. 553-562 A NEW PROOF OF THE SMOOTHNESS OF 4-POINT DESLAURIERS-DUBUC SCHEME YOUCHUN TANG, KWAN PYO KO* AND BYUNG-GOOK LEE Abstract. It is well-known
More informationPoint Lattices in Computer Graphics and Visualization how signal processing may help computer graphics
Point Lattices in Computer Graphics and Visualization how signal processing may help computer graphics Dimitri Van De Ville Ecole Polytechnique Fédérale de Lausanne Biomedical Imaging Group dimitri.vandeville@epfl.ch
More informationRemoving a mixture of Gaussian and impulsive noise using the total variation functional and split Bregman iterative method
ANZIAM J. 56 (CTAC2014) pp.c52 C67, 2015 C52 Removing a mixture of Gaussian and impulsive noise using the total variation functional and split Bregman iterative method Bishnu P. Lamichhane 1 (Received
More informationResearch on the Image Denoising Method Based on Partial Differential Equations
BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 16, No 5 Special Issue on Application of Advanced Computing and Simulation in Information Systems Sofia 2016 Print ISSN: 1311-9702;
More informationBeyond Wavelets: Multiscale Geometric Representations
Beyond Wavelets: Multiscale Geometric Representations Minh N. Do Department of Electrical and Computer Engineering University of Illinois at Urbana-Champaign www.ifp.uiuc.edu/ minhdo minhdo@uiuc.edu Acknowledgments
More informationEdges, interpolation, templates. Nuno Vasconcelos ECE Department, UCSD (with thanks to David Forsyth)
Edges, interpolation, templates Nuno Vasconcelos ECE Department, UCSD (with thanks to David Forsyth) Edge detection edge detection has many applications in image processing an edge detector implements
More informationDesign of High-Performance Filter Banks for Image Coding
Design of High-Performance Filter Banks for Image Coding Di Xu Michael D. Adams Dept. of Elec. and Comp. Engineering University of Victoria, Canada IEEE Symposium on Signal Processing and Information Technology,
More informationComputer Vision 2. SS 18 Dr. Benjamin Guthier Professur für Bildverarbeitung. Computer Vision 2 Dr. Benjamin Guthier
Computer Vision 2 SS 18 Dr. Benjamin Guthier Professur für Bildverarbeitung Computer Vision 2 Dr. Benjamin Guthier 1. IMAGE PROCESSING Computer Vision 2 Dr. Benjamin Guthier Content of this Chapter Non-linear
More informationTernary Butterfly Subdivision
Ternary Butterfly Subdivision Ruotian Ling a,b Xiaonan Luo b Zhongxian Chen b,c a Department of Computer Science, The University of Hong Kong b Computer Application Institute, Sun Yat-sen University c
More informationA NEW ROBUST IMAGE WATERMARKING SCHEME BASED ON DWT WITH SVD
A NEW ROBUST IMAGE WATERMARKING SCHEME BASED ON WITH S.Shanmugaprabha PG Scholar, Dept of Computer Science & Engineering VMKV Engineering College, Salem India N.Malmurugan Director Sri Ranganathar Institute
More informationRelationship between Fourier Space and Image Space. Academic Resource Center
Relationship between Fourier Space and Image Space Academic Resource Center Presentation Outline What is an image? Noise Why do we transform images? What is the Fourier Transform? Examples of images in
More informationBayesian Spherical Wavelet Shrinkage: Applications to Shape Analysis
Bayesian Spherical Wavelet Shrinkage: Applications to Shape Analysis Xavier Le Faucheur a, Brani Vidakovic b and Allen Tannenbaum a a School of Electrical and Computer Engineering, b Department of Biomedical
More informationScientific Computing: Interpolation
Scientific Computing: Interpolation Aleksandar Donev Courant Institute, NYU donev@courant.nyu.edu Course MATH-GA.243 or CSCI-GA.22, Fall 25 October 22nd, 25 A. Donev (Courant Institute) Lecture VIII /22/25
More informationDesign of Orthogonal Graph Wavelet Filter Banks
Design of Orthogonal Graph Wavelet Filter Banks Xi ZHANG Department of Communication Engineering and Informatics The University of Electro-Communications Chofu-shi, Tokyo, 182-8585 JAPAN E-mail: zhangxi@uec.ac.jp
More informationFiltering Images. Contents
Image Processing and Data Visualization with MATLAB Filtering Images Hansrudi Noser June 8-9, 010 UZH, Multimedia and Robotics Summer School Noise Smoothing Filters Sigmoid Filters Gradient Filters Contents
More informationImage denoising in the wavelet domain using Improved Neigh-shrink
Image denoising in the wavelet domain using Improved Neigh-shrink Rahim Kamran 1, Mehdi Nasri, Hossein Nezamabadi-pour 3, Saeid Saryazdi 4 1 Rahimkamran008@gmail.com nasri_me@yahoo.com 3 nezam@uk.ac.ir
More informationSubdivision Surfaces
Subdivision Surfaces 1 Geometric Modeling Sometimes need more than polygon meshes Smooth surfaces Traditional geometric modeling used NURBS Non uniform rational B-Spline Demo 2 Problems with NURBS A single
More informationStationary and Non-Stationary Univariate Subdivision Schemes
Punjab University Journal of Mathematics (ISSN 6-56) Vol. 5(3)(8) pp. 5- Stationary and Non-Stationary Univariate Subdivision Schemes Muhammad Asghar Department of Mathematics, The Islamia University of
More informationDepartment of Mathematics and Computer Science University of Missouri St. Louis St. Louis, MO 63121
Triangular 7 and Quadrilateral 5 Subdivision Schemes: Regular Case Charles K. Chui 1, Qingtang Jiang 2 3, Rokhaya N. Ndao Department of Mathematics and Computer Science University of Missouri St. Louis
More informationMULTI-RESOLUTION STATISTICAL ANALYSIS ON GRAPH STRUCTURED DATA IN NEUROIMAGING
MULTI-RESOLUTION STATISTICAL ANALYSIS ON GRAPH STRUCTURED DATA IN NEUROIMAGING, Vikas Singh, Moo Chung, Nagesh Adluru, Barbara B. Bendlin, Sterling C. Johnson University of Wisconsin Madison Apr. 19, 2015
More informationIEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 14, NO. 12, DECEMBER Minh N. Do and Martin Vetterli, Fellow, IEEE
IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 14, NO. 12, DECEMBER 2005 2091 The Contourlet Transform: An Efficient Directional Multiresolution Image Representation Minh N. Do and Martin Vetterli, Fellow,
More informationDenoising and Edge Detection Using Sobelmethod
International OPEN ACCESS Journal Of Modern Engineering Research (IJMER) Denoising and Edge Detection Using Sobelmethod P. Sravya 1, T. Rupa devi 2, M. Janardhana Rao 3, K. Sai Jagadeesh 4, K. Prasanna
More informationCHAPTER 6. 6 Huffman Coding Based Image Compression Using Complex Wavelet Transform. 6.3 Wavelet Transform based compression technique 106
CHAPTER 6 6 Huffman Coding Based Image Compression Using Complex Wavelet Transform Page No 6.1 Introduction 103 6.2 Compression Techniques 104 103 6.2.1 Lossless compression 105 6.2.2 Lossy compression
More informationContourlets and Sparse Image Expansions
Contourlets and Sparse Image Expansions Minh N. Do Department of Electrical and Computer Engineering University of Illinois, Urbana IL 61801 ABSTRACT Recently, the contourlet transform 1 has been developed
More informationImplementation of Lifting-Based Two Dimensional Discrete Wavelet Transform on FPGA Using Pipeline Architecture
International Journal of Computer Trends and Technology (IJCTT) volume 5 number 5 Nov 2013 Implementation of Lifting-Based Two Dimensional Discrete Wavelet Transform on FPGA Using Pipeline Architecture
More informationError-control vs classical multiresolution algorithms for image compression
Error-control vs classical multiresolution algorithms for image compression S. AMAT, J.C. TRILLO AND P. VIALA Departamento de Matemática Aplicada y Estadística. Universidad Politécnica de Cartagena. Paseo
More informationExample: Loop Scheme. Example: Loop Scheme. What makes a good scheme? recursive application leads to a smooth surface.
Example: Loop Scheme What makes a good scheme? recursive application leads to a smooth surface 200, Denis Zorin Example: Loop Scheme Refinement rule 200, Denis Zorin Example: Loop Scheme Two geometric
More informationSubdivision overview
Subdivision overview CS4620 Lecture 16 2018 Steve Marschner 1 Introduction: corner cutting Piecewise linear curve too jagged for you? Lop off the corners! results in a curve with twice as many corners
More informationFinal Review. Image Processing CSE 166 Lecture 18
Final Review Image Processing CSE 166 Lecture 18 Topics covered Basis vectors Matrix based transforms Wavelet transform Image compression Image watermarking Morphological image processing Segmentation
More informationCOMPACTLY SUPPORTED TIGHT AFFINE SPLINE FRAMES IN L 2 (R d ) AMOS RON AND ZUOWEI SHEN
MATHEMATICS OF COMPUTATION Volume 67, Number 1, January 1998, Pages 191 07 S 005-5718(98)00898-9 COMPACTLY SUPPORTED TIGHT AFFINE SPLINE FRAMES IN L (R d ) AMOS RON AND ZUOWEI SHEN Abstract. The theory
More informationAn Interface-fitted Mesh Generator and Polytopal Element Methods for Elliptic Interface Problems
An Interface-fitted Mesh Generator and Polytopal Element Methods for Elliptic Interface Problems Long Chen University of California, Irvine chenlong@math.uci.edu Joint work with: Huayi Wei (Xiangtan University),
More informationImplicit Matrix Representations of Rational Bézier Curves and Surfaces
Implicit Matrix Representations of Rational Bézier Curves and Surfaces Laurent Busé INRIA Sophia Antipolis, France Laurent.Buse@inria.fr GD/SPM Conference, Denver, USA November 11, 2013 Overall motivation
More informationScattered data reconstruction by regularization in B-spline and associated wavelet spaces
Scattered data reconstruction by regularization in B-spline and associated wavelet spaces Michael J. Johnson, Zuowei Shen and Yuhong Xu Abstract The problem of fitting a nice curve or surface to scattered,
More informationarxiv: v1 [math.na] 20 Sep 2016
arxiv:1609.06236v1 [math.na] 20 Sep 2016 A Local Mesh Modification Strategy for Interface Problems with Application to Shape and Topology Optimization P. Gangl 1,2 and U. Langer 3 1 Doctoral Program Comp.
More informationECE 533 Digital Image Processing- Fall Group Project Embedded Image coding using zero-trees of Wavelet Transform
ECE 533 Digital Image Processing- Fall 2003 Group Project Embedded Image coding using zero-trees of Wavelet Transform Harish Rajagopal Brett Buehl 12/11/03 Contributions Tasks Harish Rajagopal (%) Brett
More informationTensor products in a wavelet setting
Chapter 8 Tensor products in a wavelet setting In Chapter 7 we defined tensor products in terms of vectors, and we saw that the tensor product of two vectors is in fact a matrix. The same construction
More informationParameterization of triangular meshes
Parameterization of triangular meshes Michael S. Floater November 10, 2009 Triangular meshes are often used to represent surfaces, at least initially, one reason being that meshes are relatively easy to
More informationECS 178 Course Notes REFINEMENT
ECS 78 Course Notes REFINEMENT Kenneth I Joy Institute for Data Analysis and Visualization Department of Computer Science University of California, Davis Overview Bézier curves, B-spline curves and subdivision
More informationChapter 1 Introduction Motivation Approach Significance of Research Overview of Material... 5
ACKNOWLEDGEMENT First and foremost, I would like to thank my thesis advisor, Dr. A. Lynn Abbott, for his invaluable guidance and patience throughout my entire study at Virginia Tech. Beyond gaining academic
More informationSubdivision on Arbitrary Meshes: Algorithms and Theory
Subdivision on Arbitrary Meshes: Algorithms and Theory Denis Zorin New York University 719 Broadway, 12th floor, New York, USA E-mail: dzorin@mrl.nyu.edu Subdivision surfaces have become a standard geometric
More informationNumerical Analysis I - Final Exam Matrikelnummer:
Dr. Behrens Center for Mathematical Sciences Technische Universität München Winter Term 2005/2006 Name: Numerical Analysis I - Final Exam Matrikelnummer: I agree to the publication of the results of this
More informationReversible Wavelets for Embedded Image Compression. Sri Rama Prasanna Pavani Electrical and Computer Engineering, CU Boulder
Reversible Wavelets for Embedded Image Compression Sri Rama Prasanna Pavani Electrical and Computer Engineering, CU Boulder pavani@colorado.edu APPM 7400 - Wavelets and Imaging Prof. Gregory Beylkin -
More informationTHE problem of finding efficient representations of images. Directionlets: Anisotropic Multidirectional Representation With Separable Filtering
1916 IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 15, NO. 7, JULY 2006 Directionlets: Anisotropic Multidirectional Representation With Separable Filtering Vladan Velisavljević, Student Member, IEEE, Baltasar
More informationGeneralized trace ratio optimization and applications
Generalized trace ratio optimization and applications Mohammed Bellalij, Saïd Hanafi, Rita Macedo and Raca Todosijevic University of Valenciennes, France PGMO Days, 2-4 October 2013 ENSTA ParisTech PGMO
More informationMedical Image Compression Using Wavelets
IOSR Journal of VLSI and Signal Processing (IOSR-JVSP) Volume, Issue 4 (May. Jun. 03), PP 0-06 e-issn: 39 400, p-issn No. : 39 497 Medical Image Compression Using Wavelets K Gopi, Dr. T. Rama Shri Asst.
More informationNatural image modeling using complex wavelets
Natural image modeling using complex wavelets André Jalobeanu Bayesian Learning group, NASA Ames - Moffett Field, CA Laure Blanc-Féraud, Josiane Zerubia Ariana, CNRS / INRIA / UNSA - Sophia Antipolis,
More information