So it's for a car rental company or something to track how far the cars are driven?
Is that why you were light on details? Is this another case of a company too cheap to hire a competent engineer so they come to the Arduino board to get it done for free? Will this product be released open source?
The reason I'm asking is because I don't want to share the OCR code if it is going in something proprietary or for some company. If it is a hobby thing or open source invention type thing then I'll share code. But if this is for your employer then this is not something I can help with.
Coming to the open source community to get your engineering done for free is a bit like showing up at the homeless soup kitchen to get soup to sell at your restaurant. That's wrong on so many levels. I hope that's not who you are.
OCR is pretty much out of scope of Arduinos, mostly due to the very limited memory. That's why the discussion very quickly moved on to Arduino-friendly alternatives to OCR to read the display (because you asked how to read the display rather than how to do OCR).
I'd say just go with a Raspberry Pi or other SBC. It's probably possible to do this with an Arduino board which uses a more powerful microcontroller than the ATmega2560 but really this is an application where a computer is more suitable.
That's a good scenario, or any scenario where odometer readings are explicitly required. My particular use case is for automating the collection of this data to speed up filing paperwork for travel reimbursement working for a certain employer. In the course of the day, this employee may make several trips spread apart throughout the day, with non reimbursable driving potentially in between.
To add another level of complexity, there are different rates for different tasks. However, if I can solve the core logic of the scenario I laid out, tracking different kinds of trips shouldn't be a problem on my end. There are only two kinds of trips with two reimbursement rates.
pert:
I'd say just go with a Raspberry Pi or other SBC. It's probably possible to do this with an Arduino board which uses a more powerful microcontroller than the ATmega2560 but really this is an application where a computer is more suitable.
Thank you for your reply. Good insight. I'm leaning toward raspi at this point because it seems I really will need the extra processing and ram.
@Delta_G I will release my final project as open source. It will be available for anyone who wants to use it for free. My goal is two fold here: I would like to get some experience working on a somewhat complicated project (and I do enjoy this sort of thing). Additionally I hope it will provide some utility for someone I know. I will be presenting it to them as a gift when I am done (which will probably take some time). Because I will be using masks whoever uses it for another model of car where the odometer will be different and the steering column will not be positioned the same (perhaps), the masks will have to be done car model by car model. But it will be available for anyone that wants to create their own "masks". I intend to make a video from what I learn to make this easier for someone that hasn't done it before.
There is a 3rd intention as well. I am a 4th year student software developer and I am looking to expand my portfolio.