|
This article is cited in 4 scientific papers (total in 4 papers)
Pattern Recognition
Analysis of properties of dyadic patterns for the fast Hough transform
S. M. Karpenkoab, E. I. Ershovb a Moscow Institute of Physics and Technology (State University), Moscow, Russia
b Kharkevich Institute for Information Transmission Problems,
Russian Academy of Sciences, Moscow, Russia
Abstract:
We obtain an estimate for the maximum deviation from a geometric straight line to a discrete (dyadic) pattern approximating this line which is used for computing the fast Hough transform (discrete Radon transform) for a square image with side $n=2^p$, $p\in\mathbb{N}$. For $p$ even, the maximum deviation amounts to ${p}/{6}$. An important role in the proof is played by analysis of subtle properties of a simple combinatorial object, an array of cyclic shifts of an arbitrary binary number.
Keywords:
fast Hough transform, fast Radon transform, dyadic pattern, error analysis, combinatorial optimization, binary words.
Received: 04.07.2017 Revised: 30.07.2021 Accepted: 07.08.2021
Citation:
S. M. Karpenko, E. I. Ershov, “Analysis of properties of dyadic patterns for the fast Hough transform”, Probl. Peredachi Inf., 57:3 (2021), 102–111; Problems Inform. Transmission, 57:3 (2021), 292–300
Linking options:
https://www.mathnet.ru/eng/ppi2350 https://www.mathnet.ru/eng/ppi/v57/i3/p102
|
Statistics & downloads: |
Abstract page: | 134 | Full-text PDF : | 4 | References: | 16 | First page: | 17 |
|