Frank Nordberg writes:
| I've started building a multiformat sheet music search engine, indexing
| music in GIF, JPEG, PS, PDF and ABC formats (the only truly
| cross-platform compatible formats for notated music). So far I've only
| managed to catalogue 47 sites (less than one tenth of the sites that are
| actually out there) and the stats are:
|
| GIF/JPEG: 7560
| PDF/PS: 1740
| ABC: 6056
So how does one go about identifying which of the graphics files
and/or PDF/PS files contain music? For example, if my search programs
finds a text/whatever file called MistCoveredMountain.abc, it's not
too difficult for a program to read it and decipher the ABC inside,
recognize that it is ABC, and extract useful information.
I could easily use abc2ps and my conversion scripts to produce:
MistCoveredMountain.ps
MistCoveredMountain.pdf
MistCoveredMountain.gif
MistCoveredMountain.png
...
It seems to me that discovering that these contain music is far from
trivial. For the gif and png files, it's probably not possible with
the current state of AI. The file name is evocative, but it is much
more likely to be a picture than a tune.
How did you discover those 7560 images of music? How would one go
about sifting through the billions of images on the Net and
identifying the ones that are images of music?
To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html
