RecognizeText(Printed) is not recognizing the pound symbol (£)
I have many cases of pictures of texts where one can find a pound sign (£) but the sign is NEVER correctly recognized by Azure Cognitive Services RecognizeText API, as far as I tested. Other symbols, like the dollar sign ($) for example, are identified without problems.
I made tests with print screens of texts containing £, since these should be easy for the OCR tool to convert, and again the pound sign is not correctly identified (it becomes an f, a 2, a 1, a $ etc).
I am suspecting that the pound sign is not included in the set of characters that the tool supports, although I couldn't find a specific mention of that in the documentation (only that the tool is experimental and is optimized for English).
Has anyone been able to correctly convert a £ using the tool, or does anyone know FOR SURE (possibly through documentation) that £ is not included in their character set?
I attached one example of a very easy image which is correctly parsed except for the pound symbol £.
One thought: sometimes, completely ignoring the existence of the symbol would be better than parsing it as something else. For example, when parsing the correct price £2,00, it would be OK to receive 2,00 but converting £ to 1 (it has happened with me) makes the price change to 12,00, which completely changes the content of the information.
We are having the same problem, never have we found the pound symbol. Mostly we get E, f, $, L, which we are handling, but the problem is when we get 2, 6, 5 instead of the pound symbol as get an incorrect amount (£55.00 converted as 655.00). Really interested on this matter.
Elisio Quintino commented
I wonder if anyone from Microsoft's team had the time to take a look at this. It has been one month without activity.
Best regards, Elisio