Evolving estimators of the pointwise Hoelder exponent with Genetic Programming

Created by W.Langdon from gp-bibliography.bib Revision:1.8010

@Article{Trujillo201261,

• author = "Leonardo Trujillo and Pierrick Legrand and Gustavo Olague and Jacques Levy-Vehel",
• title = "Evolving estimators of the pointwise Hoelder exponent with Genetic Programming",
• title2 = "Evolving estimators of the pointwise Holder exponent with Genetic Programming",
• journal = "Information Sciences",
• volume = "209",
• pages = "61--79",
• year = "2012",
• ISSN = "0020-0255",
• DOI = "doi:10.1016/j.ins.2012.04.043",
• URL = "http://www.sciencedirect.com/science/article/pii/S0020025512003386",
• URL = "http://hal.inria.fr/hal-00643387",
• URL = "http://hal.inria.fr/docs/00/64/33/87/PDF/INS-S-11-01794-extrait.pdf",
• language = "ENG",
• oai = "oai:hal.inria.fr:hal-00643387",
• keywords = "genetic algorithms, genetic programming, Hoelder regularity, Local image description",
• abstract = "The regularity of a signal can be numerically expressed using Hoelder exponents, which characterise the singular structures a signal contains. In particular, within the domains of image processing and image understanding, regularity-based analysis can be used to describe local image shape and appearance. However, estimating the Hoelder exponent is not a trivial task, and current methods tend to be computationally slow and complex. This work presents an approach to automatically synthesise estimators of the pointwise Hoelder exponent for digital images. This task is formulated as an optimisation problem and Genetic Programming (GP) is used to search for operators that can approximate a traditional estimator, the oscillations method. Experimental results show that GP can generate estimators that achieve a low error and a high correlation with the ground truth estimation. Furthermore, most of the GP estimators are faster than traditional approaches, in some cases their run time is orders of magnitude smaller. This result allowed us to implement a real-time estimation of the Hoelder exponent on a live video signal, the first such implementation in current literature. Moreover, the evolved estimators are used to generate local descriptors of salient image regions, a task for which a stable and robust matching is achieved, comparable with state-of-the-art methods. In conclusion, the evolved estimators produced by GP could help expand the application domain of Hoelder regularity within the fields of image analysis and signal processing.",
• notes = "Entered for 2013 HUMIES GECCO 2013",

}

Genetic Programming entries for Leonardo Trujillo Pierrick Legrand Gustavo Olague Jacques Levy-Vehel

Citations