I recognize this situation. top-down view onto a bulk carrier ship that carries bulk goods (grain, powders, …). the big fat straw is part of the harbor infrastructure. it sucks the bulk up and sends it onwards on land.
monocular vision can do nothing in this situation.
you appear to be new to computer vision/machine vision. you’d have to learn an entire lecture’s worth of material and then spend months working on a prototype.
just engage some engineering company to do this for you.