Options

[under-score]

It seems that options are not recognized, when trying
options = list(classify_bln_numeric_mode = "0")

[eipi10]

I'm having this issue as well. For example, when I try x = ocr(img, options=list(textord_tabfind_find_tables="1")) (in order to try and OCR a table from and image), I get the following error:

Error in ocr(img, options = list(textord_tabfind_find_tables = "1")) : 
  unused argument (options = list(textord_tabfind_find_tables = "1"))

Would it be possible to add an example or two to the help for ?ocr to show how to use the options argument? The help page has a link to the tesseract engine options page, but no information about how to correctly specify these options in the ocr function.

[jeroen]

I have overhauled the way parameters are used. Can you try the dev version of tesseract and see if this works better now?

devtools::install_github("ropensci/tesseract")

[jeroen]

This should be fixed in tesseract 2.0 on CRAN. Please open an new issue if you still find problems.

[qarmitage]

I'm having what I believe is the same issue with the newest release.

`engine<-tesseract(options=list(tessedit_char_whitelist="0123456789"))
vs.
engine<-tesseract(options=list(tessedit_char_whitelist="ABCDEFGHIJKLMNOPQRSTUVWXYZ"))

cell_text <- ocr(file, engine=engine, HOCR=FALSE)`

Appear to be generating the same results.
Perhaps I'm confused.

[jeroen]

You're not confused, unfortunately this option is currently unsupported in Tesseract 4: See tesseract-ocr/tesseract#751 or tesseract-ocr/tesseract#2066

Also this wiki page: https://github.com/tesseract-ocr/tesseract/wiki/Planning#features-from-30x-which-are-missing-for-lstm

Hopefully the tesseract people will fix this soon.

[sowla]

Thank you for a really awesome package! Bartłomiej Uliasz says here:

To use whitelist in a config file or using the -c tessedit_char_whitelist=... command-line switch, in the newest 4.0 version you will have to set OCR Engine mode to the "Original Tesseract only".

I don't know if it applies to the R version or how difficult to it is to implement, but I stumbled across the post and thought I'd share it here in case someone who can judge whether this information is useful might see it :)

[jeroen]

The whitelist option is now supported again in Tesseract 4.1 (on CRAN now).