tesseract_download: Tesseract Training Data

Description

Helper function to download training data from the official tessdata repository. On Linux, the fast training data can be installed directly with yum or apt-get.

Helper function to download training data from the contributed tessdata_contrib repository.

Usage

tesseract_download(
  lang,
  model = c("fast", "best"),
  datapath = NULL,
  progress = interactive()
)
tesseract_contributed_download(
  lang,
  model = c("fast", "best"),
  datapath = NULL,
  progress = interactive()
)

Value

no return value, called for side effects

Arguments

lang: three letter code for language, see tessdata repository.
model: either fast or best is currently supported. The latter downloads more accurate (but slower) trained models for Tesseract 4.0 or higher
datapath: destination directory where to download store the file
progress: print progress while downloading

Details

Tesseract uses training data to perform OCR. Most systems default to English training data. To improve OCR performance for other languages you can to install the training data from your distribution. For example to install the spanish training data:

tesseract-ocr-spa (Debian, Ubuntu)
tesseract-langpack-spa (Fedora, EPEL)

On Windows and MacOS you can install languages using the tesseract_download function which downloads training data directly from github and stores it in a the path on disk given by the TESSDATA_PREFIX variable.

References

tesseract wiki: training data

Examples

Run this code

# download the french training data
# this is wrapped around a \donttest{} block because otherwise the clang19
# CRAN check will fail with a "> 5 seconds" message
# \donttest{
 dir <- tempdir()
 tesseract_download("fra", model = "best", datapath = dir)
 file <- system.file("examples", "french.png", package = "cpp11tesseract")
 text <- ocr(file, engine = tesseract("fra", datapath = dir))
 cat(text)
# }
# download the greek training data
# this is wrapped around a \donttest{} block because otherwise the clang19
# CRAN check will fail with a "> 5 seconds" message
# \donttest{
 dir <- tempdir()
 tesseract_contributed_download("grc_hist", model = "best", datapath = dir)
 file <- system.file("examples", "polytonicgreek.png",
   package = "cpp11tesseract")
 text <- ocr(file, engine = tesseract("grc_hist", datapath = dir))
 cat(text)
# }

Run the code above in your browser using DataLab

Last chance! 50% off unlimited learning