[OSM-talk] Pictures of opening hours signs for machine learning purposes

Toggenburger Lukas Lukas.Toggenburger at fhgr.ch
Mon Apr 5 11:02:52 UTC 2021


> What I found was that Tesseract was far too poor at recognizing the text to be useful.

It was also one of the student's finding that Tesseract did not perform very well.

> It would be great to use the photos you have as a starting point to see what level of coverage is feasible with a larger sample size.

Great to see that someone is interested! How shall we proceed? Can you provide some kind of online storage where I can upload the files? Would you like to start with only a subset so that you can check on feasibility?

Lukas


More information about the talk mailing list