Pin It

Follow Us On Twitter

Featured Videos

New computer code spots fake artwork

sparseResearchers have developed a new way to identify fake and imitation artwork known as ‘sparse coding.’

In short, the new methods creates a virtual library of known works by an artist and then breaks down each paintings visual elements to identify their style.  Artwork that is identifiable will contain the same simple elements while imitation works will not.

The study of artworks using mathematical analysis is a new subject area that gained attention when it was found in 1999 that many of the ‘drip paintings’ by Jackson Pollock were created in patterns that existed in fractal mathematics.

There is still controversy however about whether the same mathematical analysis could actually identify works by the prominent artist.

Over the last few years many scientists have taken many different approaches to delineating the true origin of several pieces of art, each with different results.

However, Daniel Rockmore and colleagues at the US Dartmouth College now believe that sparse coding may be in fact the most veritable technique yet known.

‘Sparse coding’ takes digital blocks of a piece of art dividing a piece into 144 squares of 12 rows by 12 columns.  Then each set is given black and white forms and assigned a pattern.

Each piece is then modified by a computer until the pieces can be combined proportionally to create the piece in question.  The smallest possible set of details that can be used to create a piece are known as the ‘sparset’ which identifies an artist’s general painting style and asserts the validity of a painting.


Leave a Reply




You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>