|
This article is cited in 3 scientific papers (total in 3 papers)
IMAGE PROCESSING, PATTERN RECOGNITION
Morphlets: a new class of tree-structured morphological descriptors of image shape
Yu. V. Vizilter, V. S. Gorbatcevich, S. Yu. Zheltov, A. Yu. Rubis, A. V. Vorotnikov FGUP «GosNIIAS»
Abstract:
In this paper, we present a new class of tree-structured morphological descriptors of image shape. The notion of a Haar-like function is formalized. It is demonstrated that Pytiev morphology of centered images can be described in terms of a Haar-like basis. The morphology of localized mosaic shapes and corresponding localized Haar-like wavelet systems are described. The formal way for generating a hierarchical structure of localized mosaic shapes and corresponding tree of localized Haar-like wavelets is reported. The notion of “morphlets” as Haar-like wavelets corresponding to the mosaic shape of some image is introduced. Some constructive schemes for the generation of morphlet-based image descriptors are considered.
Keywords:
image analysis, mathematical morphology, Pytiev morphology, Wavelets, Shape comparison.
Received: 03.12.2014
Citation:
Yu. V. Vizilter, V. S. Gorbatcevich, S. Yu. Zheltov, A. Yu. Rubis, A. V. Vorotnikov, “Morphlets: a new class of tree-structured morphological descriptors of image shape”, Computer Optics, 39:1 (2015), 101–108
Linking options:
https://www.mathnet.ru/eng/co63 https://www.mathnet.ru/eng/co/v39/i1/p101
|
Statistics & downloads: |
Abstract page: | 270 | Full-text PDF : | 99 | References: | 40 |
|