Improve OCR recognition/processing of ingredients

[AcuarioCat]

The OCR is pretty good but a high proportion of the unknown ingredients are as a result of very minor errors such as a missing comma between ingredients or common characters such as 'o' and 'a', 'i' and 'l' and 'rn (R N)' and 'm' being mis-interpreted.

I don't know if there is any processing between the extraction of words from the image and the words added to the list of ingredients but if there is then this would be a perfect place to pre-analyse the words.

As the language is generally known then running the word through the ingredients analysis for that language would determine if it was recognized or not, Substitutions of letters and attempts with split words (in the case of missing commas) could then be made to see if a match could be found. As the errors are fairly consistent this could be a fairly rapid process.

Part of

• Add more ingredient analysis (tracker) #5706

[aleene]

I am not sure, but I think the latest spelling improvements are now tested out in french only. Check out #spelling on Slack.

[teolemon]

@rbournhonesque is currently reprocessing old OCRs. This should help.

[github-actions]

This issue has been open 90 days with no activity. Can you give it a little love by linking it to a parent issue, adding relevant labels and projets, creating a mockup if applicable, adding code pointers from https://github.com/openfoodfacts/openfoodfacts-server/blob/main/.github/labeler.yml, giving it a priority, editing the original issue to have a more comprehensive description… Thank you very much for your contribution to 🍊 Open Food Facts