Stories
Slash Boxes
Comments

SoylentNews is people

posted by martyb on Saturday March 17 2018, @01:56AM   Printer-friendly
from the Count-Craters?-There's-no-Duke-or-Earl? dept.

Using AI to count craters on the moon at U of T's Centre for Planetary Sciences

A new technique developed by researchers at the University of Toronto is using the technology behind self-driving cars to measure the size and location of crater impacts on the moon.

"When it comes to counting craters on the moon, it's a pretty archaic method," says Mohamad Ali-Dib, a postdoctoral researcher at the Centre for Planetary Sciences (CPS) at U of T Scarborough.

"Basically we need to manually look at an image, locate and count the craters and then calculate how large they are based off the size of the image. Here we've developed a technique from artificial intelligence that can automate this entire process that saves significant time and effort."

[...] In order to determine its accuracy, the researchers first trained the neural network on a large data set covering two-thirds of the moon, and then tested their trained network on the remaining third of the moon. It worked so well that it was able to identify twice as many craters as traditional manual counting. In fact, it was able to identify about 6,000 previously unidentified craters on the moon.

Also at New Scientist and Science News.

Lunar Crater Identification via Deep Learning (arXiv:1803.02192)


Original Submission

 
This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: -1, Offtopic) by Anonymous Coward on Saturday March 17 2018, @02:34AM (5 children)

    by Anonymous Coward on Saturday March 17 2018, @02:34AM (#653912)

    Cicked the link. Landed on the correct page. Read the story. Then looked at the links on the page. Sorry, University of Toronto leaves a lot to be desired. I'm left wondering what their true mission is. With a name like "University", you would suspect that they are in the business of education. The links all seem to have something to do with "Trump is a failure" and "diversity".

    Maybe the staff at U of T should take back their website. And, maybe they should get back to the business of education.

    • (Score: 0, Offtopic) by Ethanol-fueled on Saturday March 17 2018, @02:43AM (4 children)

      by Ethanol-fueled (2792) on Saturday March 17 2018, @02:43AM (#653915) Homepage

      Ho-ho, God. You're right. From the summary link I clicked on the Future Students [utoronto.ca] link because that's probably the best path to witness the extent of the indoctrination.

      Sure enough, all Chinks and 2 Pajeets with the only Whites being the fat Hipster Soyboy bitch and the perfect skinny blonde woman that all the foreigners go for. Now, Foreign men do love their blonde White women. Asians, Latinos, Blacks, Pajeets.

      I guess Pink-haired dykes with no bras and three layers of tit-fat as ideologies don't look so good on paper after all.

      • (Score: -1, Offtopic) by Anonymous Coward on Saturday March 17 2018, @02:54AM (2 children)

        by Anonymous Coward on Saturday March 17 2018, @02:54AM (#653919)

        How the hell did they check their work? Maybe they went to the moon and physically counted craters to double check the results? Maybe some of those craters aren't even there, but have been extrapolated by the algorithm. To summarize the story, "Mohammed and his guesswork algorithm supposes that he can find twice as many craters on the moon than other people have found."

        • (Score: 1) by Ethanol-fueled on Saturday March 17 2018, @03:41AM (1 child)

          by Ethanol-fueled (2792) on Saturday March 17 2018, @03:41AM (#653940) Homepage

          They have their 1 or 2 Pajeets/Akhmeds who believe in the cause, who are good at their work. Nice Blackboard scrawlings, in America blackboard scrawlings in the background show everybody else how smart you are. That's why in America Big Bang Theory is somehow popular. We all have a select few of those productive types.

          But the text reads like garbage. Perhaps the University of Toronto had better hire some of their Marketing/Communications graduates to clean things up. And if they already did, go study someplace else.

          • (Score: 0) by Anonymous Coward on Saturday March 17 2018, @02:04PM

            by Anonymous Coward on Saturday March 17 2018, @02:04PM (#654065)

            That really sucks man. I did my thesis on blackboard scrawlings. I get paid a lot of money to do background scrawling for movies, newscasts, and even for academia. Now, you're telling me that some pajeets and some ragheads are moving in on my territory? I'll have to look into it.

      • (Score: -1, Offtopic) by Anonymous Coward on Saturday March 17 2018, @03:31AM

        by Anonymous Coward on Saturday March 17 2018, @03:31AM (#653937)

        Pink-haired dykes

        I'm beginning to believe there are only two types of women:

        1. As you mentioned
        2. Bisexuals who ride the cock carousel

  • (Score: 3, Funny) by stormwyrm on Saturday March 17 2018, @04:27AM (6 children)

    by stormwyrm (717) on Saturday March 17 2018, @04:27AM (#653952) Journal
    One crater! Two craters! Three craters! ... Hahahahaha! (cue thunder)
    --
    Numquam ponenda est pluralitas sine necessitate.
    • (Score: 2) by frojack on Saturday March 17 2018, @05:41AM (5 children)

      by frojack (1554) on Saturday March 17 2018, @05:41AM (#653967) Journal

      So humans had to count 2/3rds of all craters and teach the AI what was and was not a crater. It would have been easier just to count the other third.

        How did the AI do when presented with random selection of craters from Mars or some other moon?
      And now that we know.....?

      --
      No, you are mistaken. I've always had this sig.
      • (Score: 2) by maxwell demon on Saturday March 17 2018, @08:57AM (3 children)

        by maxwell demon (1608) on Saturday March 17 2018, @08:57AM (#654001) Journal

        If they just had counted the craters by hand, you probably would never have heard about it.

        For the data itself, it probably would have sufficient to extrapolate the last third. I don't think a crater more or less makes a big difference in whatever they are going to use that data for.

        --
        The Tao of math: The numbers you can count are not the real numbers.
        • (Score: 0) by Anonymous Coward on Saturday March 17 2018, @09:47AM (2 children)

          by Anonymous Coward on Saturday March 17 2018, @09:47AM (#654010)

          "There are 4000 holes in the road in Blackburn Lancashire, one twenty-sixth of a hole per person, according to a council survey. If Blackburn is typical then there are over two million holes in Britain's roads and 300 000 in London." (attributed to the 17 January 1967 Daily Mail)

          • (Score: 2) by maxwell demon on Saturday March 17 2018, @10:10AM (1 child)

            by maxwell demon (1608) on Saturday March 17 2018, @10:10AM (#654014) Journal

            You know the difference between 2/3 and 1/500? Moreover I'm sure the scientists know how to select an unbiased sample, which the Daily Mail sample obviously was not.

            --
            The Tao of math: The numbers you can count are not the real numbers.
            • (Score: 0) by Anonymous Coward on Saturday March 17 2018, @10:24PM

              by Anonymous Coward on Saturday March 17 2018, @10:24PM (#654239)

              Whoosh!

      • (Score: 1, Insightful) by Anonymous Coward on Saturday March 17 2018, @09:19AM

        by Anonymous Coward on Saturday March 17 2018, @09:19AM (#654005)

        > So humans had to count 2/3rds of all craters and teach the AI what was and was not a crater. It would have been easier just to count the other third.

        That was my first thought as well, but then I read the next sentence :) and found out that the student had surpassed its masters:

        > It worked so well that it was able to identify twice as many craters as traditional manual counting. In fact, it was able to identify about 6,000 previously unidentified craters on the moon.

        (I assume that they later checked that those were actual craters and not a figment of AI's imagination.)

  • (Score: 2) by The Mighty Buzzard on Saturday March 17 2018, @11:29AM (2 children)

    by The Mighty Buzzard (18) Subscriber Badge <themightybuzzard@proton.me> on Saturday March 17 2018, @11:29AM (#654024) Homepage Journal

    I hope they realize they're taking food from the mouths of innocent illegal aliens by doing this. And making the task more expensive to boot. Half a dozen illegals from the Home Depot parking lot, a large map, and some colored pencils would be vastly less expensive and would put money in the hands of the people instead of corporate interests.

    --
    My rights don't end where your fear begins.
(1)