Driver's License Scanner

A real-time web application that scans and decodes PDF417 barcodes on driver's licenses using computer vision and multiple barcode detection libraries.

How It Works

Barcode Detection Process

1. Camera Feed: Captures live video from selected camera
2. Scan Area: Defines a centered rectangular region (70% width × 15% height) for PDF417 detection
3. Multi-Library Approach: Uses multiple detection methods in priority order:
   • ZXing-CPP: Primary decoder with excellent PDF417 support
   • OpenCV: Backup barcode detector
   • pyzbar: Fallback with multiple image preprocessing techniques

ANSI Data Parsing

The PDF417 barcode contains ANSI-encoded driver's license data with standardized field codes:

• DCS: Last Name
• DCT/DAC: First Name
• DBD: Issue Date
• DBA: Expiration Date
• DBB: Date of Birth
• DBC: Sex
• DAY: Eye Color
• DAU: Height
• DAG: Address
• DAI: City
• DAJ: State
• DAK: ZIP Code
• DAQ: License Number

Image Processing Pipeline

For each frame, the system:

1. Extracts the defined scan area
2. Applies multiple preprocessing techniques (grayscale, binary threshold, morphological operations)
3. Attempts barcode detection with each method
4. Parses successful reads into human-readable format
5. Returns both prettified data and raw barcode content

Setup

Prerequisites

• Python 3.7+
• Webcam or camera device (iPhone or high-quality cameras recommended for dense PDF417 barcodes)
• Modern web browser with camera permissions

Installation

1. Clone/Download the project files

2. Install Dependencies:

pip install -r requirements.txt

sudo apt-get install libzbar0 libzbar-dev

Usage

Privacy & Statistics

Important: This scanner tracks successful scans for statistical purposes only. Each successful scan increments a counter that can be accessed via the /scan-stats API endpoint. No personal data from licenses is stored - only scan counts and timestamps are recorded. Example: GET https://dl-scanner.chrisccluk.live/scan-stats

Starting the Application

python app.py        # Default port 5000
python app.py 8080   # Custom port 8080

The server starts at http://localhost:5000 (or your specified port)

Scanning Process

1. Select Camera: Choose from available cameras in the dropdown
2. Start Camera: Click "Start Camera" to begin video feed
3. Customize Overlay: Adjust box width/height percentages or use URL parameters (?width=80&height=20)
4. Position License: Place driver's license so the back barcode is within the green alignment box
5. Automatic Scanning: The app continuously scans every second
6. Results: When successful, displays:
   • Annotated image showing detected barcode area
   • Prettified license information (copy-pasteable)
   • Collapsible raw barcode data with copy button
   • Download option for annotated image

Example Photos

Publicly generated photo (NOT REAL)

Tips for Best Results

• Camera Quality: Use iPhone or high-quality cameras - PDF417 barcodes contain dense information requiring good resolution
• Lighting: Ensure even, bright lighting without shadows or glare
• Positioning: Keep the PDF417 barcode fully within the green box
• Orientation: Use the flip camera button if the barcode appears mirrored
• Stability: Hold the license steady during scanning
• Distance: Maintain appropriate distance for clear barcode visibility
• Angle: Keep the license flat and parallel to the camera

Features

• Real-time Detection: Continuous scanning without manual capture
• Multiple Decoders: Fallback detection methods for reliability
• Camera Controls: Flip camera orientation for better positioning
• Scan Statistics: Persistent counter tracking successful scans
• Data Export: Copy prettified or raw data to clipboard
• Image Download: Save annotated scan results
• Responsive Design: Works on desktop and mobile browsers
• Security: No server-side file storage, base64 image embedding
• Privacy: Only scan statistics stored, no personal data retention
• Session Management: Automatic idle user redirection to prevent resource consumption

Troubleshooting

No Detection:

• Improve lighting conditions
• Ensure barcode is within green alignment box
• Try different camera angles
• Check if ZXing-CPP is installed for better detection

Camera Issues:

• Grant camera permissions in browser
• Try different cameras from the dropdown
• Refresh page and restart camera

Poor Image Quality:

• Clean camera lens
• Increase lighting
• Reduce camera shake
• Move closer/farther for optimal focus

Technical Details

• Backend: Flask web server with persistent scan statistics
• Computer Vision: OpenCV for image processing
• Barcode Libraries: ZXing-CPP, pyzbar, OpenCV barcode detector
• Frontend: Vanilla JavaScript with WebRTC camera access
• Data Storage: JSON file for scan counter persistence
• Security: Client-side image handling, no server file storage