Low-level Edge Detection Using Genetic Programming: performance, specificity and application to real-world signals

Created by W.Langdon from gp-bibliography.bib Revision:1.9101

@TechReport{Harris:1997:ledGPpsa,

• author = "Christopher Harris and Bernard Buxton",
• title = "Low-level Edge Detection Using Genetic Programming: performance, specificity and application to real-world signals",
• year = "1997",
• institution = "UCL",
• type = "Research Note",
• number = "RN/97/7",
• address = "Gower Street, London, WC1E 6BT, UK",
• URL = "http://www.cs.ucl.ac.uk/staff/W.Langdon/ftp/papers/rn_97_7.pdf",
• broken = "http://citeseer.ist.psu.edu/404512.html",
• keywords = "genetic algorithms, genetic programming, Edge Detection, convolution filter, Canny, Spacek, cubic spline, Petrou",
• size = "23 pages",
• abstract = "...we show that [GP]... can produce detectos that often out perform the established 'optimal' detectors...",
• notes = "

  404512.html PDF link broken 22 Oct 2004

  spline = x*x*x +2*x*x +x

  datasets: underwater, office, synthetic step edge, theoretical

  p18 'data driven approach to creating customised edge detectors for use within a specific domain. Such detectors have improved performance and high specificy, reflecting the characteristics of the data set used to drive the evolution.'

  Oct 2018 rn_97_7.pdf is scan of photocopy...",

}

Genetic Programming entries for Christopher Harris Bernard Buxton

Citations