Truck load identification

edit

Hi, I am working on a problem using IOT sensors etc to identify how much load is filled in a truck.example 30% truck is empty or filled with air. I am exploring if this problem can be solved with some accuracy using CV algorithms where I can train models by taking pictures with various filled capacities and then try to deduct actual fulfillment given set of pictures.

Let me know if this is the right approach to solve it or any other approach should I use.

Thanks for your guidance, Pritam