Filtering the image archive…

Home Forums Babel Image Archives Filtering the image archive…

This topic contains 2 replies, has 2 voices, and was last updated by  Brandon 5 days, 2 hours ago.

Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • #23385 Reply

    Peter Kilmartin

    is there no way, using software, that we could use computers to look at each image and discard all the ones that look like fuzz. i imagine a software application that ‘looks’ at each picture through some sort of averaging algorithm (something that would do the same job as screwing up your eyes and blurring the image) to discard all the useless ones.
    this has got to be possible and it could be achieved using cloud computing, nasa did an experiment that required you to leave your PC on 24 hours a day and they used the processing power while you were not (i.e. when the screen saver was on).
    Also.. is there a way of getting each image from the archive to load faster, that way we have more chance of seeing something… anything…

    #23390 Reply

    Delengroth
    Participant

    This would all be possible if Jonathan releases the source code. He said he would a number of years ago, but I think he wanted to add some extra features first, such as being able to search multiple pages or entire books. Personally, I don’t care too much for that feature, and it might even be easier to write it if the source was open for everyone to try. Until then, there’s not much that can be done other than reverse engineering his exact algorithm (so that your locations match the ones here).

    #23524 Reply

    Brandon

    What if you could also implement object detection , so if an image is found with an object , it is tagged. A average object detection program can run from 100-250 images per second.

    The likleyhood of finding something normal is small, but it would be interesting to see what would get tagged and what wouldn’t

Viewing 3 posts - 1 through 3 (of 3 total)
Reply To: Filtering the image archive…
Your information: