Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Generation of fractals using the Burt pyramid

Not Accessible

Your library or personal account may give you access

Abstract

Fractals have been used successfully to describe complex, irregular natural forms such as clouds, coastlines, trees, and mountains, which are difficult to characterize in terms of continuous, differentiable functions.1 One of the most important properties of fractals is their self-similarity. As the spatial scale of viewing changes over many orders of magnitude, the fractal remains similar in structure and appearance. The Burt pyramid2 decomposes an image into a sum of octave width spatial frequency bands plus a low-pass filtered component. Local spatial information is also preserved. If an inherently self-similar fractal is represented in pyramid form, one would expect each spatial band to look similar to the others. Conversely, if similar patterns are entered into each spatial band of a pyramid, the sum should be a self-similar fractallike texture. We have used this technique to generate realistic looking textures including clouds, water waves, galaxies, woodgrain, and rock. The time required to synthesize a fractallike texture is much reduced over present methods, suggesting that this method may be useful for computer artwork and animation. A real-time animation sequence of clouds and waves is shown. This sequence was produced on a DeAnza image processor driven by a PDP-11/24.

© 1985 Optical Society of America

PDF Article
More Like This
Manipulating image information with pyramids

Edward H. Adelson and Peter J. Burt
THX2 OSA Annual Meeting (FIO) 1985

Diffusion and Reaction in Tortuous Structures as Modelled by Fractals

J. Klafter and J. M. Drake
WB3 Microphysics of Surfaces, Beams, and Adsorbates (MSBA) 1985

The use of fractals in optical wave propagation through random media

Bruce J. West
CThK1 Conference on Lasers and Electro-Optics (CLEO:S&I) 1992

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.