On the theory of function-valued mappings and its application to the processing of hyperspectral images

详细信息    查看全文

• 作者：Daniel Otero^a ; ^{dotero@uwaterloo.ca} ; Davide La Torre^b ; ^c ; ^{davide.latorre@unimi.it} ; ^{davide.latorre@nu.edu.kz} ; Oleg Michailovich^d ; ^{olegm@uwaterloo.ca} ; Edward R. Vrscay^a ; ^{ervrscay@uwaterloo.ca}
• 关键词：Function-valued functions ; Image processing ; Banach spaces ; Fourier transform ; And fractal transform
• 刊名：Signal Processing
• 出版年：2017
• 出版时间：May 2017
• 年：2017
• 卷：134
• 期：Complete
• 页码：185-196
• 全文大小：1568 K
• 卷排序：134

文摘

We propose an alternative approach for the analysis of multidimensional images. Fourier transforms of function-valued mappings (FVMs) are defined. We derive an Euler-Lagrange equation for functionals which operate on FVMs. A class of fractal transforms (FTs) for FVMs is defined, allowing fractal coding. We show how an FT on an FVM induces a generalized FT on its Fourier transform.