Convert JPEG to PDF with invisible text on image

[Cmoutier]

Hello,

I want to use PaddleOCR for replacment of ABBYY Fine Reader. i need to convert multiple JPEG files to one PDF with invisible text on image. I want to search on select text on my generated PDF file.
Is it possible with PaddleOCR?

Thanks a lot

[GreatV]

Yes, it is possible to use PaddleOCR to convert multiple JPEG files into a single PDF with invisible text over the image, ensuring that the text can be selected and searched within the generated PDF.

Steps to achieve this:

1. Extract Text Using PaddleOCR:

   • Use PaddleOCR to recognize and extract text from the JPEG files.
   • Ensure you preprocess images properly to improve OCR accuracy (e.g., contrast adjustment, deskewing).

2. Generate a PDF with OCR Layer:

   • Use a Python library like reportlab or Pillow to create a PDF containing the original image.
   • Overlay the extracted text in an invisible layer (using the pdfplumber or PyMuPDF libraries).
   • The text should be placed at the coordinates where PaddleOCR detected them in the image.

3. Combine Multiple Pages into One PDF:

   • Convert each processed JPEG with overlay text into a PDF page.
   • Merge all pages into a single PDF using PyPDF2 or pdfplumber.

Tools Required:

• PaddleOCR (for text extraction)
• Pillow (for image processing)
• reportlab/PyPDF2/pdfplumber/PyMuPDF (for creating PDF with invisible text)

Would you like a code example to achieve this?

---

Response generated by 🤖 feifei-bot | chatgpt-4o-latest

⚠️ This is an AI-generated response. Please verify any critical information.

[Cmoutier]

Thanks for your answer!

I have tried that with canvas, but my results are not good. It's better to use pdfplumber or PyMuPDF?

We can't give several JPEG fils to PaddleOCR to generate directly a single PDF, instaed of generate a lot of PDF files to merge after?

I have a lot of JPEG files to convert....