Hello everyone, I'm working on a project based on maze solving with image processing.
Whole process works like that in the simulation : Simulation video
However, the problem I'm facing is: I want my drone to move center of the maze. However, thresholding the image taken by drone inversely, then moving on the white area works on simulation but not in real.
As you can see in this image there are so many other black and white objects. Under effect of irregular lighting conditions, it becomes really hard to distinguish colors of the maze from other substances, even with adaptive thresholding method.
I don't have any problem with solving the maze. Birefly, I want my drone to move the center of the maze autonomously. If you guys offer me any idea or solution I'll really be appreciated.
I'm using python-opencv