PDFOutliner – Intelligent Document Structure Extraction

📑 Overview

PDFOutliner is a Python-based tool that extracts clean, hierarchical outlines from PDF documents.
By analyzing document structure, it generates a navigable JSON representation of headings (H1–H3) and their relationships, enabling advanced navigation, search, and content analysis.

---

✨ Key Features

🧠 Intelligent Structure Recognition

• Analyzes font size, weight, and positioning to detect headings
• Identifies hierarchical relationships between headings
• Preserves original document organization

⚙️ Robust Processing Engine

• Works with inconsistently formatted PDFs
• Supports multi-page documents
• Language-agnostic text analysis

📂 Clean, Structured Output

• Standards-compliant JSON output
• Preserves heading hierarchy
• Includes page numbers for easy navigation

---

🚀 Getting Started

📋 Prerequisites

• Python 3.10 or later
• pip (Python package manager)
• Docker (optional, for containerized usage)

---

💻 Installation (Python)

# Clone the repository
git clone https://github.com/Dhwanitisshah/Structured-PDF-Outline-Extractor.git
cd Structured-PDF-Outline-Extractor

# Install dependencies
pip install -r requirements.txt

---

🐳 Installation (Docker)

# Build the Docker image
docker build -t pdfoutliner .

# Run the container
docker run -v "$(pwd)/input:/app/input" -v "$(pwd)/output:/app/output" pdfoutliner

---

📊 Usage

Python

# Place your PDF in the input folder
mkdir -p input output
cp your_document.pdf input/

# Run the script
python pdf_outliner.py

# Output will be in the 'output' folder as JSON

Docker

docker run -v "$(pwd)/input:/app/input" -v "$(pwd)/output:/app/output" pdfoutliner

---

📥 Input / 📤 Output

Input

• Location: /app/input
• Format: PDF files (*.pdf)

Output

• Location: /app/output
• Format: JSON (same name as input PDF)

Example Output

{
  "title": "Sample Document",
  "headings": [
    {
      "text": "Introduction",
      "level": 1,
      "page": 1,
      "children": [
        {
          "text": "Background",
          "level": 2,
          "page": 1
        }
      ]
    }
  ]
}

---

🛠 Technical Details

Architecture – Hybrid approach combining:

1. Native PDF Structure Analysis – Extracts built-in outlines when available
2. Heuristic Font Analysis – Detects headings based on font size, weight, and style
3. Position-Based Context – Uses text positioning to infer hierarchy

Dependencies

• PyMuPDF – Layout-aware text extraction
• pdfplumber – Font metadata analysis
• Python Standard Libraries – re, collections, json, logging

---

📂 Project Structure

Structured-PDF-Outline-Extractor/
│── input/                # PDF files to process
│── output/               # Extracted JSON outlines
│── pdf_outliner.py       # Main script
│── requirements.txt      # Python dependencies
│── Dockerfile            # Docker build configuration
│── README.md             # Project documentation

---

🔮 Future Development

• 🌐 Enhanced Language Support for non-Latin scripts
• 🤖 NLP Integration for better heading context analysis
• 📊 Confidence Scoring for extraction accuracy
• 🌍 Web API for remote PDF processing
• 📦 Batch Processing for multiple files

---

📄 License

This project is licensed under the MIT License – see the LICENSE file for details.