Computer program developed to detect art forgery

The Chronicle of Higher Education reports researchers have devel­oped a way to use a com­puter and image soft­ware to ana­lyze a paint­ing and deter­mine whether it is gen­uine or a forgery.

The exper­i­ment was per­formed on 101 high-resolution scans of Van Gough paint­ings from muse­ums in the Netherlands. The pro­gram ana­lyzed the artist’s brush strokes and cre­ate algo­rithm to describe Van Gough’s style.

The researchers told the CHE that while the algo­rithm was pretty suc­cess­ful, it could still be con­fused by the vari­ety of brush strokes that the artist would use in a sin­gle painting.

An algo­rithm to describe the way an artist paints. It reminds me of the idea that a mil­lion mon­keys sit­ting at a mil­lion type­writ­ers will even­tu­ally pro­duce Hamlet; it’s one of those attempts to ratio­nal­ize artis­tic expres­sion, to quan­tify it and explain it. All the bet­ter to sell it with, I suppose.

Share and Enjoy:
  • Print
  • Diigo
  • Digg
  • del.icio.us
  • Facebook
  • Google Bookmarks
  • Posterous
  • LinkedIn
  • Ping.fm
  • Tumblr

Related posts:

  1. Tagging Art
  2. Video Games as Art
  3. Journalist Bailout Program
  4. Computer in the shop: Day 3
  5. Whither rock ‘n roll?
This entry was posted in Authority Issues and tagged , , , , . Bookmark the permalink. Both comments and trackbacks are currently closed.
  • Recent Comments