News for hackers. Stuff that matters.
Re:Examples to compare OCR services: Amazon vs. Google vs. M... by bduerst on Thursday July 18, 2019 @03:00PM (#20472929)
Would have been nice to see % accuracy as well as or instead of just % words matched. It seems like the author decided to forego false positives.

You can train models that are good at finding words in images but can have terrible accuracy at perceiving the actual text of the word.

Re:Examples to compare OCR services: Amazon vs. Google vs. M... by ocrcustomserver on Friday July 19, 2019 @07:24PM (#20483246)
Typically OCR accuracy is measured in two ways, CER (Character error rate) and WER (Word error rate). If just one number is provided, it's typically CER.

"Finding words in images" is a bit ambiguous. It can mean "word spotting" (intended for retrieval rather than transcription) but also "text segmentation" (part of preprocessing step before OCR).

Re:Examples to compare OCR services: Amazon vs. Google vs. M... by rhizome on Thursday July 18, 2019 @09:02PM (#20475425)
What distinction are you making? I think it might be a little too subtle written that way.