mirror/tesseract
1
0
Fork 0
mirror of https://github.com/tesseract-ocr/tesseract.git synced 2024-11-27 20:59:36 +08:00
Code Issues Actions 11 Packages Projects Releases Wiki Activity

Limit unicharset from training_text only to Indic languages

Browse Source
This commit is contained in:
Shree Devi Kumar 2021-03-14 17:58:57 +00:00
parent a589ded25f
commit efe9ff611f
1 changed files with 9 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
src/training/tesstrain_utils.sh
View File

@ -349,11 +349,17 @@ phase_I_generate_image() {
# Phase UP : Generate (U)nicharset and (P)roperties file.
phase_UP_generate_unicharset() {
tlog "\n=== Phase UP: Generating unicharset and unichar properties files from Training Text ==="
tlog "\n=== Phase UP: Generating unicharset and unichar properties files ==="
local box_files=$(ls ${TRAINING_DIR}/*.box)
UNICHARSET_FILE="${TRAINING_DIR}/${LANG_CODE}.unicharset"
run_command unicharset_extractor --output_unicharset "${UNICHARSET_FILE}" \
--norm_mode "${NORM_MODE}" ${TRAINING_TEXT}
if [[ "${NORM_MODE}" == "2" ]] && [[ "${LANG_IS_RTL}" == "0" ]] ; then
run_command unicharset_extractor --output_unicharset "${UNICHARSET_FILE}" \
--norm_mode "${NORM_MODE}" ${TRAINING_TEXT}
else
run_command unicharset_extractor --output_unicharset "${UNICHARSET_FILE}" \
--norm_mode "${NORM_MODE}" ${box_files}
fi
check_file_readable ${UNICHARSET_FILE}
XHEIGHTS_FILE="${TRAINING_DIR}/${LANG_CODE}.xheights"