Having decided to at least try using OCR, and Tesseract specifically, I need to decide what I'm going to run my new machine on. Throw in the requirement that, if all goes well, I will have to take a picture of the State Transition Table of my larger machine to apply the OCR to, and I'm coming up Raspberry Pi. It doesn't hurt that Pis are well understood and supported by the maker community. A quick search for "tesseract and raspberry pi" turns up a lot of relevant howtos on making this work.
Thinking ahead, my new 6-state / 6-symbol Turing machine will need a way to present the Tape. Since the cells can now contain six (maybe seven ;-) different symbols, a mechanical solution is pretty much out of the question (sorry Flip-Bit). So that leaves 7-segment displays or an LCD of some sort. I'm opting for the Official Raspberry Pi 7" Touchscreen LCD. It's been well reviewed, has for sure hassle-free integration with the Pi, and the touchscreen will give me a lot of flexibility when creating a UI.
So my target runtime environment will be a Raspberry Pi, Official 7" Touchscreen LCD, and a Raspberry Pi Camera module. I'm looking forward to really working with the Pi. It's not like I don't have a bunch of them already running OctoPi, OSMC, OpenMediaVault, and MAME. It's just that those are all distributions that are easily installed and setup. Parts are ordered.
Getting back to running Tesseract on the Pi, all of the howtos suggest using the following software stack:
That sounds good to me. I have never done Python development before so that will be fun.
Since all of the above software is cross platform I have decided to at least start development on my Windows machine. I have installed my old friend Eclipse (the Liclipse distribution) with all of the aforementioned software. Ready to kick some OCR ass.