Amit Dovev

Did you try both the `fast` and `best` models?

If you know in advance that the text is written in a specific language, like English in this case, another thing you can try is to add the parameter `-l...

You can try this command: `convert image.png -bordercolor White -border 4x4 image-4-4-wb.png`

`-l +equ` This is wrong usage. Drop the `+equ` `-l ` and to turn on the equation detection feature add this to the command: `-c textord_equation_detect=true` In any case, the...

https://github.com/tesseract-ocr/tessdata_fast/issues/4#issuecomment-572740279 (referring to the omission of `equ.traineddata` from the `tessdata_fast` repo) jbreiden commented on 9 Jan 2020 > I just noticed this issue. The equation recognizer was left out deliberately...

You can still try to detect equations using the correct usage. At least with `--oem 0` (the old ocr engine) and a lang.traineddata from the `tessdata` repo, it might sometimes...

@callegar, I accidentally overrode your comment with my reply. Sorry! I fixed it. Your comment was restored.

Related issue: #3513.

You can try with a model that contains lstm and legacy data: https://github.com/tesseract-ocr/tessdata/raw/main/deu.traineddata

>Error: Tesseract (legacy) engine requested, but components are not present in /usr/share/tesseract-ocr/5/tessdata/deu.traineddata!! You told Tesseract to use the legacy engine but the traineddata model in your tessedata directory does not...