使用基样条构造多分辨率逼近及小波

我们首先介绍了Ｂ－样条及基样条，然后用ｍ阶的Ｂ－样条Ｎｍ（ｘ）生成一个Ｌ＾２（Ｒ）中一个比例为ｒ的多分辨逼近，而且用（ψｔ（ｘ）＝Ｌ＾（ｍ）２ｍ（ｒｘ－ｔ），ｔ＝１，２，．．．ｘ－１）构造了相应的小波空间，这里Ｌ２ｍ为２ｍ阶的基样条，最后，我们给出了小波的分解与合成算法。     

L~2([0,1])的半正交小波基及其对偶小波基

本文从样条函数出发利用折叠法得到了L2（[0,1]）空间的两组相互对偶的半正交小波基，这两组小波基有显式表达式，并导出了他们的小波分解与重建算法．进一步，应用这些小波基给出了刻划高阶Holder空间的一个充分条件．     

多小波子空间上的单小波表示

本文在较弱的条件下,建立了2重多小波子空间与单小波子空间的关系.即由2重多小波构造出单小波.一方面,这种单小波的平移伸缩与2重多小波的平移伸缩生成的子空间是完全相同的;另一方面,它具有插值性.因此通过构造出的单小波建立了多小波子空间上的Shannon型采样定理.     

具有矩阵伸缩的双正交小波基

在这篇文章里,我们研究了伸缩为矩阵的双正交小波基的构造问题,在适当条件下,我们得到了Ｌ~２（Ｒ~ｎ）的小波框架或双正交小波基｛ｓｊ,ｋ｝和｛ｓｊ,ｋ｝,其中ｓｊｋ（ｘ）＝ｄｅｔＡｓ（Ａｊｘ－ｋ）,ｓｊ,ｋ（ｘ）＝ｄｅｔＡｊ２ｓ（Ａｊｘ－ｋ）（ｊ Ｚ． ｋ Ｚ~ｎ）及 Ａ是一伸缩矩阵．     

小波框架的某些判定准则

本文首先给出了希尔伯特空间H上两个半小波框架序列成为小波框架的一个充分条件,该结论与框架的扰动理论有关.然后建立了通过膨胀与平移母函数生成L~2(R)上的小波框架的膨胀参数,平移参数以及母函数的一些充分条件.这些结果推广了小波框架理论中经典文献中的相关结论.我们还对贝塞尔序列进行了讨论,并得到了一些有趣的结论.     

小波分析方法及其应用

２．正交小波和多分辨分析前面已经指出,连续小波变换和离散小波变换具有统一的形式,特别是正交小波的引入,使一个小波函数的“伸缩”和“平移”产生的函数族构成函数空间Ｌ２（Ｒ）的一个标准正交基,这给信号分析和一般的数据处理带来许多方便。这样就产生一个问题：...     

L^2[0,1]上的正交小波基

从尺度因子Ｍ＝４的正交小波基出发,利用折叠方法得到了Ｌ＾２「０,１」空间的正交小波基,这种小波不同于折叠前的小波基,它是完全限制在有限区间「０,１」上,且保持小波基的正交性,并在使用过程中拥有更大的灵活性。也可用类似方法对一般尺度小波进行折叠。     

Hardy空间H~1(R)的紧小波框架系数刻画

利用标准正交小波基下函数的展开系数来刻画Hardy空间H~1(R)已经得到了很好的证明.该文利用紧小波框架与标准正交小波基的关系及其性质,给出了Hardy空间H~1(R)在紧小波框架下函数展开系数的一个刻画.     

M-带插值小波包

本文给出M-带插值小波包的构造.M-带插值小波包是根据基插值函数建立的迭代函数序列进行伸缩平移的空间序列.这种小波包可使信号分解更为精细,并具有更好的局部性.由此建立了这种小波包子空间上的近似采样定理.     

正交小波包的构造

本文给出尺度因子ａ＝４时正交小波包的构造，推广了［２，４］引入的正交小波包，并给出相应的分解与再构造算法．本文引入的正交小波包具有保持信号ｆ∈Ｌ^２的线性相位，也讨论了尺度因子ａ＝ｋ（ｋ∈Ｚ，ｋ≥２）正交小波的构造     

Classification of Wavelet Bases by Translation Subgroups and Nonharmonic Wavelet Bases

Abstract The structure of the set S of shiftable points of wavelet subspaces is researched in this paper. We prove that S = ℝ or where q∈ℕ. The spectral and functional characterizations for the shiftability are given. Furthermore, the nonharmonic wavelet bases are discussed. This work is supported by Postdoctoral Research Fund of China, NSF of China under Grant 69772025 and Open Fund of National Laboratory for Machine Perception of Peking University     

一种具有消失矩的小波基的构造

在N≥2情况下构造了一种性质良好的小波母函数,这种小波母函数N阶可导并且趋于零的阶数为O|t|-N(|t|→∞),同时具有N-2阶消失矩和某种对称性.文章还就N=4的情况给出了计算实例.     

Haar Type Orthonormal Wavelet Bases in R2

K.-H. Grochenig and A. Haas asked whether for every expanding integer matrix A ∈ M_n(ℤ) there is a Haar type orthonormal wavelet basis having dilation factor A and translation lattice ℤⁿ. They proved that this is the case when the dimension n = 1. This article shows that this is also the case when the dimension n = 2.     

Interpolatory Band-Limited Wavelet Bases on the Sphere

In the present paper, we construct space-localized bases for the space $W_n^n:=\oplus_{k=n+1}^{2n} Harm_k({\Bbb S}^2)$ of band-limited functions on the sphere. Each of the basis functions is a zonal polynomial centered at a point $\eta_i\in{\Bbb S}^2$. The goal of this work is to describe explicit fundamental systems $\lbrace\eta_j\rbrace_{j=1,\dots,M_n}$ for the space $W_n^n$ which finally lead to space- and frequency-localized polynomial bases for $L^2({\Bbb S}^2)$.     

Quantifying Democracy of Wavelet Bases in Lorentz Spaces

We study the efficiency of the greedy algorithm for wavelet bases in Lorentz spaces in order to give the near best approximation. The result is used to give sharp inclusions for the approximation spaces in terms of discrete Lorentz sequence spaces.     

Anisotropic Wavelet Bases and Restricted Nonlinear Approximation

Restricted nonlinear approximation is a generalization of n‐term approximation in which a weight function is used to control the terms of the approximant. Here, restricted nonlinear approximation is considered with respect to anisotropic wavelet bases. In particular, characterizations for those functions, which provide a specific convergence rate by restricted nonlinear approximation, are presented.     

A Framework for Image Deblurring Using Wavelet Packet Bases

We show in this paper that the average over translations of an operator diagonal in a wavelet packet basis is a convolution. We also show that an operator diagonal in a wavelet packet basis can be decomposed into several operators of the same kind, each of them being better conditioned. We investigate the possibility of using such a convolution to approximate a given convolution (in practice an image blur). Then we use these approximations to deblur images. First, we show that this framework permits us to redefine existing deblurring methods. Then, we show that it permits us to define a new variational method which combines the wavelet packet and the total variation approaches. We argue and show by experiments that this permits us to avoid the drawbacks of both approaches which are, respectively, the ringing and the staircasing.     

Spectral Analysis of the Differential Operator in Wavelet Bases

We analyze the spectral properties of the differential operator in wavelet bases. The problem is studied on a periodic domain, with periodized wavelets. An algorithm for finding the eigenvalue function of the differential operator is presented, and general conditions that ensure a "nicely behaving" eigenvalue function are derived.     

Adelic Multiresolution Analysis, Construction of Wavelet Bases and Pseudo-Differential Operators

In our previous paper, the Haar multiresolution analysis (MRA) $\{V_{j}\}_{j\in \mathbb {Z}}$ in $L^{2}(\mathbb {A})$ was constructed, where $\mathbb {A}$ is the adele ring. Since $L^{2}(\mathbb {A})$ is the infinite tensor product of the spaces $L^{2}({\mathbb {Q}}_{p})$ , p=∞,2,3,…, the adelic MRA has some specific properties different from the corresponding finite-dimensional ones. Nevertheless, this infinite-dimensional MRA inherits almost all basic properties of the finite-dimensional case. In this paper we derive explicit formulas for bases in V _j , $j\in \mathbb {Z}$ , and for the wavelet bases generated by the above-mentioned adelic MRA. In view of the specific properties of the adelic MRA, there arise some technical problems in the construction of wavelet bases. These problems were solved with the aid of the operator formalization of the process of generation of wavelet bases. We study the spectral properties of the fractional operator introduced by S. Torba and W.A. Zúñiga-Galindo. We prove that the constructed wavelet functions are eigenfunctions of this fractional operator. This paper, as well as our previous paper, introduces new ideas to construct different infinite-dimensional MRAs. Our results can be used in the theory of adelic pseudo-differential operators and equations over the ring of adeles and in adelic models in physics.