Hi!
I am working on a Robot Vision project using OpenCV. First up I am working on a maze navigation problem and am trying to get the robot to identify the walls of the maze and then make a decision on which direction to move.
So far I am making reasonable progress using downloaded images and video and can detect the lines that I want (e.g. where the wall meets the floor). The lines in the image and video are detected using canny edge detection and the lines are further defined using HougLinesP and then using a loop the lines I want to focus on are highlighted using a loop to draw on the lines that have been found.
Now I am moving to using the robots onboard camera - Raspberry Pi Camera.
I have amended the image / video capture to now use the Pi Camera, first problem was that it was rotated 180 degrees (upside down) which I have now fixed.
I can view the original video being captured and also the canny edge detection version of the image. I have now hit some problems in that I cannot highlight the lines using the same for loop as it returns various errors.
I believe the problem has something to do with either the masked region of interest that I am setting and the resolution and image size of the new image.
Processor is a Raspberry Pi 3 and I am using Python
Here is a link to my code on GitHub:
Thank you so much in advance for your help!