Feedback Latin.traineddata

[TheSeiko]

trafficstars

All Errors are in comparison to the old Latin.traineddata: https://github.com/tesseract-ocr/tessdata/issues/82

Compiled 13.09.2017 tesseract Windows 10 x86

---

Recognition Error A -> Ä

1454419179964_2.txt.txt

[TheSeiko]

Problems with E/F at begin of line Often "E" is recognised as "F"

1454436670744_subImage.txt

[TheSeiko]

E/F

1454438383366_subImage.txt

[TheSeiko]

small/capital letters

1454455689122_subImage.txt

[TheSeiko]

When a sentence spans multiple lines and the only one or two words are in the last line, the words in last line are often recognised with capital letters. Here: Wird Expected: wird

[stweil]

I just tried all of the examples above with Tesseract 5.3.1 and tessdata_fast/script/Latin. Most of them were recognized perfectly. I only got one Ä/A and one ß/B confusion.