View : 943 Download: 0

Full metadata record

DC Field Value Language
dc.contributor.author이연주-
dc.creator이연주-
dc.date.accessioned2016-08-25T06:08:33Z-
dc.date.available2016-08-25T06:08:33Z-
dc.date.issued2007-
dc.identifier.otherOAK-000000028090-
dc.identifier.urihttps://dspace.ewha.ac.kr/handle/2015.oak/181641-
dc.identifier.urihttp://dcollection.ewha.ac.kr/jsp/common/DcLoOrgPer.jsp?sItemId=000000028090-
dc.description.abstract웨이브릿 변환은 신호, 영상처리와 수치적인 계산과 같은 다양한 응용분야에 매우 유용하다. 이 학위논문은 두 종류의 새로운 이중직교 웨이브릿 시스템의 구성과 해석에 기여한다. 첫째, 새로운 compactly supported 대칭인 이중직교 웨이브릿 시스템의 새로운 그룹을 제안한다. 이 그룹의 각각의 mask는 자유도 w를 갖는다. w=0일 때 이는 최소 길이의 이중직교 Coifman 웨이브릿 시스템이 된다. w가 0이 아닐 때 support는 약간 늘어나지만 더 좋은 smoothness 를 얻을 수 있다. 실제로 새로운 이중직교 웨이브릿 시스템이 유사 보간 세분법에서 시작됐지만 이는 Cohen, Daubechies and Feauveau가 기존에 제안한 mask와 일치했다. 그러나 대응되는 수학적 해석은 아직 이뤄지지 않았고 이 연구에서 유사 보간 세분법과 Cohen, Daubechie와 Feauveau의 mask 와의 관계를 살피고 새로운 웨이브릿 시스템의 regularity, 안정성, 선형 독립성, vanishing moments 와 정확도 등을 살펴본다. 또한, CAGD의 관점에서 유사 보간 mask 의 장점을 소개한다. 둘째, 지수 B-스프라인 에 기반을 둔 non-stationary 이중직교 웨이브릿 시스템에 대해 연구한다. refinable 함수와 웨이브릿 함수들은 모두 compactly supported이고 그 중심을 기준으로 대칭이고 이중직교 refinable 함수들은 같은 order 의 B-스프라인과 같은 regularity를 갖는다. 또한, 그 웨이브릿은 지수 함수에 대한 vanishing moments 성격을 갖고 이는 특정 주파수에 몰린 신호의 근사와 압축에 있어 효율적이다. 특히, 제안된 웨이브릿 함수들이 그들이 생성하는 함수 공간의 Riesz basis가 됨을 보이고 몇몇 예제를 통해 제안된 결과가 기존의 B-spline 보다 신호처리에 보다 효과적임을 보여준다.;The wavelet transform has been very useful in a variety of applications such as signal and image processing, and numerical computation. This thesis is devoted to constuction and analysis of two kinds of new families of biorthogonal wavelet systems. First, we present a new class of compactly supported and symmetric biorthogonal wavelet systems. Each refinement mask in this family has tension parameter ω. When ω=0, it becomes the minimal length biorthogonal Coifman wavelet system [51]. Choosing ω away from zero, we can get better smoothness of the refinable functions at the expenses of slightly larger support. Though the construction of the new biorthogonal wavelet systems, in fact, starts from a new class of quasi-interpolatory subdivision schemes, we find that the refinement masks accidently coincide with the ones by Choen, Daubechies and Feauveau[5, §6.C](or [15,§8.3.5]), which are designed for the purpose of generating biorthogonal wavelets close to orthonormal cases. However, the corresponding mathematical analysis is yet to be provided. In this study, we consider the connection between the quasi-interpolatory subdivision schemes and thd masks by Cohen, Daubechies and Feauveau, and then we study the fundamental properties of the new biorthogonal wavelet systems such as regularity, stability, linear independence, vanishing moments and accuracy. On the other hand, in view of CAGD, we present some advantage of the quasi-interpolatory mask,. Second, we are concerned with non-stationary biorthogonal wavelet systems based on exponential B-splines. The refinable functions and wavelets are compactly supported and symmetric with respect to their centers. The biorthogonal refinable functions are proved to have the same regularity as the ones of the corresponding B-spline. The wavelets have vanishing moments of exponential polynomials, which can be more effective (than polynomial vanishing moments) to signal approximation and compression if the signals are concentrated in some frequency bands. In particular, we prove that the suggested wavelets form Riesz basis for the spaces they generate. To illustrate our construction and results in this paper, we present some numerical examples which reflect more succesful applications to signal analysis than the case of the polynomial B-splines.-
dc.description.tableofcontents1 Introduction = 1 2 Preliminaries = 8 2.1 Linear independence and stability of refinable functions = 8 2.2 Biorthogonal wavelet systems = 10 2.3 Subdivision scheme = 12 3 Quasi-interpolatory refinable functions = 16 3.1 Construction of quasi-interpolatory refinable functions = 16 3.2 Linear independence and stability of refinable functions = 21 3.3 Maximal smoothness of refinable functions = 25 3.4 Asymptotic smoothness of refinable functions = 30 4 Biorthogonal multiresolution analysis and wavelets based on quasi-interpolatory refinable functions = 40 4.1 Biorthogonal wavelet systems = 40 4.2 Approximation order and vanishing moment = 44 4.3 Examples = 48 5 Non-stationary biorthogonal multiresolution analysis and wavelets based on Exponential B-spline = 51 5.1 Exponential B-splines = 51 5.2 The duals of exponential B-splines = 53 5.3 Smoothness of the dual functions = 61 5.4 Biorthogonal projections = 64 5.5 Non-stationary Riesz biorthogonal wavelets = 65 5.6 Vanishing moments with respect to exponential polynomials = 70 5.7 Non-stationary subdivision scheme = 79 6 Applications = 83 6.1 Quasi-interpolatory subdivision scheme = 83 6.2 Biorthogonal wavelet systems based on quasi-interpolatory refinable functions = 85 6.3 Non-stationary biorthogonal wavelet systems based on Exponential B-splines = 89-
dc.formatapplication/pdf-
dc.format.extent1184852 bytes-
dc.languageeng-
dc.publisher이화여자대학교 대학원-
dc.subject.ddc510-
dc.titleDevelopment of New Classes of Wavelet Systems for Data Representation-
dc.typeDoctoral Thesis-
dc.title.translated데이터 표현을 위한 새로운 웨이브릿 시스템들의 발전-
dc.creator.othernameLee, Yeon Ju-
dc.format.pageⅳ, 99 p.-
dc.identifier.thesisdegreeDoctor-
dc.identifier.major대학원 수학과-
dc.date.awarded2007. 8-
Appears in Collections:
일반대학원 > 수학과 > Theses_Ph.D
Files in This Item:
There are no files associated with this item.
Export
RIS (EndNote)
XLS (Excel)
XML


qrcode

BROWSE