Evolving approximate image filters

Simon Colton, Pedro Torres

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

11 Citations (Scopus)

Abstract

Image filtering involves taking a digital image and producing a new image from it. In software packages such as Adobe's Photoshop, image filters are used to produce artistic versions of original images. Such software usually includes hundreds of different image filtering algorithms, each with many fine-tuneable parameters. While this freedom of exploration may be liberating to artists and designers, it can be daunting for less experienced users. Photoshop provides image filter browsing technology, but does not yet enable the construction of a filter which produces a reasonable approximation of a given filtered image from a given original image. We investigate here whether it is possible to automatically evolve an image filter to approximate a target filter, given only an original image and a filtered version of the original. We describe a tree based representation for filters, the fitness functions and search techniques we employed, and we present the results of experimentation with various search setup . We demonstrate the feasibility of evolving image filters and suggest new ways to improve the process.

Original languageEnglish
Title of host publicationApplications of Evolutionary Computing - EvoWorkshops 2009
Subtitle of host publicationEvoCOMNET, EvoENVIRONMENT, EvoFIN, EvoGAMES, EvoHOT, EvoIASP, EvoINTERACTION, EvoMUSART, EvoNUM, EvoSTOC, EvoTRANSLOG, Proceedings
Pages467-477
Number of pages11
DOIs
Publication statusPublished - 23 Jul 2009
Externally publishedYes
EventEuropean Workshop on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2009 - Tübingen, Germany
Duration: 15 Apr 200917 Apr 2009
Conference number: 7th
https://link.springer.com/book/10.1007/978-3-642-01129-0 (Proceedings)

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume5484 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Workshop

WorkshopEuropean Workshop on Evolutionary and Biologically Inspired Music, Sound, Art and Design 2009
Abbreviated titleEvoMUSART 2009
Country/TerritoryGermany
CityTübingen
Period15/04/0917/04/09
Internet address

Cite this