- author = "Brian J. Ross and Han Zhu",
- title = "Procedural Texture Evolution Using Multiobjective Optimization",
- journal = "New Generation Computing",
- year = "2004",
- volume = "22",
- number = "3",
- pages = "271--293",
- month = sep,
- keywords = "genetic algorithms, genetic programming, Procedural Textures, Multi-objective Optimisation",
- ISSN = "1882-7055",
-
URL = "
http://www.cosc.brocku.ca/files/downloads/research/cs0218.pdf",
-
DOI = "
doi:10.1007/BF03040964",
- size = "23 pages",
- abstract = "investigates the application of evolutionary multi-objective optimisation to two-dimensional procedural texture synthesis. Genetic programming is used to evolve procedural texture formulae. Earlier work used multiple feature tests during fitness evaluation to rate how closely a candidate texture matches visual characteristics of a target texture image. These feature test scores were combined into an overall fitness score using a weighted sum. This paper improves this research by replacing the weighted sum with a Pareto ranking scheme, which preserves the independence of feature tests during fitness evaluation. Three experiments were performed: a pure Pareto ranking scheme, and two Pareto experiments enhanced with parameterless population divergence strategies. One divergence strategy is similar to that used by the NSGA-II system, and scores individuals using their nearest-neighbour distance in feature-space. The other strategy uses a normalised, ranked abstraction of nearest neighbour distance. A result of this work is that acceptable textures can be evolved much more efficiently and with less user intervention with MOP evolution than compared to the weighted sum approach. Although the final acceptability of a texture is ultimately a subjective decision of the user, the proposed use of multi-objective evolution is useful for generating for the user a diverse assortment of possibilities that reflect the important features of interest.",
-
notes = "cs0218.pdf refers to Brock University technical report
#CS-02-18 July 2002.
LilGP 1.1
Ohmsha Ltd",
