I currently have a dataset of x and y coordinates of shapes drawn by users on a graphics tablet.
The users are asked to copy an image by drawing it on the screen. The object they're required to copy is a collection of shapes (straight lines, squares, rectangles, triangles and circles). At the moment, I'm trying to determine the best approach to identifying individual shapes and thought that detecting lines and "snapping" them to a perfectly straight line would be a good approach and then identify corners in these shapes to determine what shapes are drawn. Is there an optimal way to detect corners and possibly snap a "wonky/wobbly" line in a matrix of x and y coordinates? Certainly not requesting someone provide the code, but more some helpful guidance as to different possible approaches and methodologies?
Thank you in advance!