Slashdot Mirror

← Back to Stories (view on slashdot.org)

Militarizing Your Backyard With Python and AI

Posted by timothy on from the it-puts-the-squirrel-in-the-bucket dept.

mikejuk writes "Kurt Grandis took some cutting edge and open source AI tools, Python, an Arduino and a SuperSoaker and built the (almost) perfect squirrel hosing machine. The project involved Open Computer Vision (OpenCV), an a SVM learning procedure that he trained to tell the difference between a squirrel and a non-squirrel. After 'perfecting' the classifier the hardware came next — a SuperSoaker Mark I was used as the 'water cannon.' A pair of servos were used to aim the gun and a third to pull the trigger."

1 of 112 comments (clear)

  1. Before you try to reproduce this... by jafo · · Score: 5, Informative

    I saw his presentation at PyCon a few weeks ago. During Q&A I asked: "My experience with OpenCV has been that it's nearly impossible to use, poor documentation, documentation of a different version of the API, build issues with the libraries. Was I just on the wrong track, or is this a common experience?"

    His answer was that it's true that it's very hard to get OpenCV working.

    Also note that after a while the squirrels stopped being annoyed by the water gun and would just sit there while getting sprayed.

    He did a very nice job of it though! I particularly like the part about using the bushy tail to tell a squirrel from a bird.