I have made a code to identify QR markers but I want to make sure that it is the correct method. My current method is to make the QR marker canonized (make the QR marker fit a square ) and then resize it to an 8 by 8 image and then see for each pixel if it is 1 or 0 and store that in an array. Then I match it with an array made from a list of the original markers. This works but some of the QR markers are not identified which is why I wanna see if there are other methods. Maybe I should look at the features of the QR markers to find a match instead?
sounds like you’re reinventing the wheel.
why not use one of the existing qr schemes in opencv, e.g. here or here ?
there’s a difference between “QR codes” and “augmented reality markers”.
QR codes convey data, and nothing else. it can be a lot of data, kilobytes even. they’re designed so that a machine can robustly recover all the information.
AR markers have a simple identity (a dozen bits or so), and they allow a camera to determine its precise location relative to the marker.
QR codes aren’t designed for localization, AR markers aren’t designed to hold a lot of data.