Install Tesseract v5.x.x and Training Data 4.1 on Amazon Linux 2023

amz2023-install-tesseract.sh

#!/usr/bin/env bash

TESSERACT_VERSION=5.3.3
TESSERACT_TRDATA_VERSION=4.1.0

TESSERACT_FILE="$TESSERACT_VERSION.tar.gz"
TESSERACT_URL="https://github.com/tesseract-ocr/tesseract/archive/$TESSERACT_FILE"

LEPTONICA_VERSION=1.84.1
LEPTONICA_FILE=leptonica-$LEPTONICA_VERSION.tar.gz
LEPTONICA_URL=https://github.com/DanBloomberg/leptonica/releases/download/$LEPTONICA_VERSION/leptonica-$LEPTONICA_VERSION.tar.gz


sudo dnf upgrade --releasever=2023.3.20240108 && \
sudo dnf groupinstall -y "Development Tools" && \
sudo dnf install -y autoconf \
  automake \
  libtool \
  libjpeg-devel \
  libpng-devel \
  giflib \
  libtiff-devel \
  zlib-devel \
  ImageMagick \
  ImageMagick-devel \
  ImageMagick-perl \
  libwebp \
  zlib \
  zlib-devel \
  libjpeg \
  libjpeg-devel \
  libwebp \
  libwebp-devel \
  libtiff \
  libpng \
  cairo \
  pango \
  libicu \
  pangomm \
  ghostscript \
  libpng-devel \
  openjpeg2-devel \
  openjpeg2-tools \
  pango-devel \
  cairo-devel \
  pango-devel \
  libicu-devel \
  libtiff-devel \
  pangomm-devel \
  libcurl-devel \
  cairomm-devel \
  pangomm-devel \
  libjpeg-turbo-devel && \
sudo yum clean all
sudo cp /usr/lib64/libjpeg.so.62 /usr/local/lib/ && \
sudo cp /usr/lib64/libwebp.so.7 /usr/local/lib/ && \
sudo cp /usr/lib64/libtiff.so.5 /usr/local/lib/ && \
sudo cp /usr/lib64/libpng16.so.16 /usr/local/lib/ && \
echo "Creating folders" && mkdir ~/libs && cd ~/libs && echo "installing leptonica..." && \
mkdir leptonica && cd leptonica && wget $LEPTONICA_URL && tar -zxvf $LEPTONICA_FILE && rm $LEPTONICA_FILE && cd leptonica-$LEPTONICA_VERSION && \
# Takes ~5 min. on T2.micro Instance machine (Free Tier).
./configure && make && sudo make install && cd ~/libs && echo 'Installing Tesseract...' && mkdir tesseract && cd tesseract
wget $TESSERACT_URL &&  tar -zxvf $TESSERACT_FILE && sudo rm $TESSERACT_FILE && cd tesseract-$TESSERACT_VERSION && \
# Takes ~10 min. on T2.micro instance (Free Tier).
./autogen.sh && ./configure && make && sudo make install && sudo ldconfig && \
echo 'Installing Training Data...' && cd /usr/local/share/tessdata && \
sudo wget https://github.com/tesseract-ocr/tessdata/raw/main/fra.traineddata && \
sudo wget https://github.com/tesseract-ocr/tessdata/raw/main/eng.traineddata
echo 'export TESSDATA_PREFIX=/usr/local/share/tessdata' >> ~/.bashrc