We can do the detecting ourselves bit by bit or using a KI model like tensorflow.
https://stackoverflow.com/questions/54082938/how-do-i-get-coordinates-of-image-shown-in-opencv
https://www.geeksforgeeks.org/find-co-ordinates-of-contours-using-opencv-python/
https://docs.opencv.org/3.4/d4/dc6/tutorial_py_template_matching.html