How could I add number plate recognition to an existing dataset?

I’m building a gadget to track my excavator bucket. It’s called ExWeigh. The object of the exercise is to weigh the material in the bucket. It also inferences:

type of bucket,

what the bucket is doing,

where it is in space,

Bucket distance (from the excavator)

number of bucket load cycles which creates a bucket count

and weight on the lift cycle. That’s pretty straight forward, well, sort of… to see what I have achieved so far follow this link to my google drive:

So I’m now trying to figure out if I can read the Rego Numbers of the truck and trailer (if in view) PLUS read the tonnage weight (an integer like → 11 or 14) normally displayed on the side of the truck and trailer, and could I incorporate this into the existing system?

Hi @automateme2!
First off, this is an incredibly interesting project and this is a perfect fit for the Roboflow platform. In fact, please reference this Roboflow blog post, How to Build an Automated License Plate Reading Application, which walks through the workflow that I recommend building for your use case. The only difference being that your workflow should include an object detection model to detect the Rego number of the truck and trailer, and the tonnage weight.**
**
To help guide your project, I have assembled a list of reference materials:

Happy building!!

Thank you!

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.