hello from sri lanka
i need help with the above
this is very new to me so please bare with me
i need to count fertilizer bags being loaded into a truck using an image processing method
i was thinking yolov8 on a rasberry pi 5 8gb that is connected to a rasberry pi camera and analog counter
the module is mounted onto a truck and the camera will face down at the entry point. any bag that is loaded by the laborer is automatically counted on a analog counter. a visual red light alarm will go if the count is achieved and not overloaded
(loading quanity is updated daily)
i have the vission on how it should work but i dont know how to get there. pleas help - i can code in python but i need a guide on the feasibility of this project
also do i need to buy the AI hat?
1 Like
Hey @monarawila297484, Welcome to the forums!
This sounds like a fun project! As long as you can create a large enough collection of photos of the fertiliser bags, you should be able to pretty consistently recognise the bags. I think the difficult part of this project is going to be deciding when exactly to count a given bag as being inside the truck.
It is likely that the model will detect, lose and then redetect a given bag as it is being loaded into the truck if the camera’s vision of the bag is obscured by the handler or if the shape of the bag changes enough when being handled to stop being detected.
I would start by having a read through our Getting Started with YOLO Object and Animal Recognition on the Raspberry Pi guide, as this will give you an idea of the basics.
You shouldn’t need to use an AI HAT for this project, a Pi 5 and a Raspberry Pi Camera Module 3 should be enough to get started although your model will perform better with the AI HAT or with the Raspberry Pi AI Camera.
Hope this helps! 
1 Like
many thanks from sri lanka
1 Like