PDF Form AutoFill System

An intelligent PDF form auto-fill system that combines OCR, computer vision, and LLM to automatically detect form fields and fill them with provided data.

Features

• Box Detection: Uses OpenCV to detect input fields in PDF forms
• OCR Text Extraction: Extracts labels using OCR.space API
• LLM Mapping: Uses Azure OpenAI to intelligently map JSON data to form fields
• Accurate Placement: Fixed coordinate system for precise text placement
• Multiple Field Types: Supports letter-by-letter and entire text filling

Setup

1. Install Dependencies:

pip install opencv-python pdf2image pytesseract PyMuPDF requests pillow

2. Configure API Keys:

   • Copy config_template.py to config.py
   • Add your API keys:
     • OCR.space API key
     • Azure OpenAI API key and endpoint

3. Run the System:

python llm_autofill_mapper.py

How It Works

1. PDF to Image: Converts PDF to image for processing
2. Field Detection: Detects input boxes using OpenCV contour detection
3. OCR Extraction: Extracts text labels using OCR.space API
4. LLM Mapping: Uses Azure OpenAI to map JSON data to detected fields
5. Text Placement: Writes text on image using correct coordinates
6. PDF Generation: Converts filled image back to PDF

Output

• output/autofilled_form.pdf - Final filled PDF
• output/filled_image.png - Filled image for verification
• output/field_mappings.json - Complete mapping details
• output/detected_all_field_types.png - Visualization of detected fields

Configuration

Edit config.py to customize:

• API keys and endpoints
• File paths
• Processing settings (DPI, font size, etc.)

Requirements

• Python 3.7+
• OpenCV
• PyMuPDF
• OCR.space API key
• Azure OpenAI API key