OCR PDF

pdf-ocr.sh

#!/bin/bash

# Ensure the required tools are installed
command -v pdftoppm >/dev/null 2>&1 || {
    echo "pdftoppm is not installed. Exiting."
    exit 1
}
command -v tesseract >/dev/null 2>&1 || {
    echo "Tesseract is not installed. Exiting."
    exit 1
}
command -v gs >/dev/null 2>&1 || {
    echo "Ghostscript is not installed. Exiting."
    exit 1
}

# Input PDF file
INPUT_PDF="$1"
OUTPUT_DIR="./ocr_output"
OUTPUT_PDF="searchable_output.pdf"

# Create a directory to store images and PDFs
mkdir -p "$OUTPUT_DIR"
echo "Converting PDF to JPG images..."

# Convert PDF to JPG images (one image per page)
pdftoppm -jpeg "$INPUT_PDF" "$OUTPUT_DIR/output"

# Process each JPG image with Tesseract to create PDFs
echo "Running OCR on images and creating individual PDFs..."
for IMG in "$OUTPUT_DIR"/*.jpg; do
    BASENAME=$(basename "$IMG" .jpg)
    OCR_PDF="$OUTPUT_DIR/$BASENAME.pdf"

    echo "Processing $IMG..."
    tesseract "$IMG" "$OUTPUT_DIR/$BASENAME" -l deu pdf # Generates searchable PDF
done

# Combine individual PDFs into a single searchable PDF using Ghostscript
echo "Merging individual PDFs into a single searchable PDF with Ghostscript..."
gs -dBATCH -dNOPAUSE -q -sDEVICE=pdfwrite -sOutputFile="$OUTPUT_PDF" "$OUTPUT_DIR"/*.pdf

echo "Process completed! Searchable PDF saved as $OUTPUT_PDF."