Tesseract-OCR quick guide

1. 安装

Github Repository: https://github.com/tesseract-ocr/tesseract

Binary Packages Download: https://digi.bib.uni-mannheim.de/tesseract/

Other installing: https://tesseract-ocr.github.io/tessdoc/Compiling.html

• 1.1 apt安装

sudo apt install tesseract-ocr
# Default installing to: /usr/share/tesseract-ocr/4.00/tessdata/

• 1.2 其他安装

centos安装（因为tesseract官方只提供 windows 和 ubuntu 的安装包），centos等只能编译安装，若不想折腾就用现成的docker镜像吧： https://github.com/tesseract-shadow/tesseract-ocr-re

docker pull tesseractshadow/tesseract4re.
docker run -dt --name t4re tesseractshadow/tesseract4re

• 1.2 检查安装

tesseract --version
tesseract --list-langs

2. Examples:

• 2.1 Importing traingdata files.
  https://github.com/tesseract-ocr/tessdata_fast

cd $TESSERACT_HOME/configs
curl -O https://raw.githubusercontent.com/tesseract-ocr/tessdata_fast/master/chi_sim.traineddata

• 2.2 Invoking image recognition:

tesseract 1.png . -l configs/chi_sim