Drug Classification System

This project processes images to extract barcodes and text, then associates them together. Perfect for drug classification and inventory management. It also includes a QR code + Medicine ID lookup system for retrieving medicine details.

Features

• Barcode Detection: Automatically detects and reads various barcode formats (QR codes, UPC, EAN, etc.)
• OCR Text Extraction: Extracts text from images using Tesseract OCR
• Association: Links barcode data with extracted text from the same image
• Batch Processing: Processes all images in a folder automatically
• JSON Output: Saves results in structured JSON format
• QR Code + Medicine ID Lookup: Upload QR code and enter Medicine ID to retrieve detailed medicine information
• Web Interface: User-friendly Flask web application for barcode and QR code lookups

Requirements

System Dependencies

• Tesseract OCR: Required for text extraction
  • macOS: brew install tesseract
  • Ubuntu/Debian: sudo apt-get install tesseract-ocr
  • Windows: Download from Tesseract GitHub

Python Dependencies

• opencv-python
• pyzbar
• pytesseract
• Pillow
• numpy

Quick Start

Option 1: Automated Setup

python setup_and_run.py

Option 2: Manual Setup

# Install Python dependencies
pip install -r requirements.txt

# Run the scanner
python barcode_scanner.py

Usage

Process All Images

python barcode_scanner.py

Process Single Image

python barcode_scanner.py --single path/to/image.jpg

Custom Options

python barcode_scanner.py --images custom_folder --output results.json

Output Format

The script generates a JSON file with the following structure:

[
  {
    "image_path": "images/sample.jpg",
    "image_name": "sample.jpg",
    "timestamp": "2024-01-01T12:00:00",
    "barcodes": [
      {
        "data": "1234567890123",
        "type": "EAN13",
        "position": {
          "x": 100,
          "y": 50,
          "width": 200,
          "height": 50
        }
      }
    ],
    "text": {
      "full_text": "Drug Name: Aspirin\nDosage: 500mg\nExpiry: 2025-12-31",
      "words": [
        {
          "text": "Aspirin",
          "confidence": 95,
          "position": {"x": 10, "y": 20, "width": 60, "height": 15}
        }
      ]
    },
    "barcode_count": 1,
    "has_text": true
  }
]

Features Explained

Barcode Detection

• Supports multiple barcode formats (QR, UPC, EAN, Code128, etc.)
• Provides barcode position information
• Handles multiple barcodes per image

Text Extraction

• Uses Tesseract OCR for high accuracy
• Extracts both full text and individual words with positions
• Includes confidence scores for each detected word

Association

• Automatically links barcodes with text from the same image
• Provides detailed position information for both barcodes and text
• Enables easy correlation between barcode data and product information

Troubleshooting

Common Issues

1. Tesseract not found

   • Install Tesseract OCR on your system
   • Make sure it's in your system PATH

2. Poor text recognition

   • Ensure images are clear and well-lit
   • Try preprocessing images for better contrast

3. No barcodes detected

   • Check if barcodes are clear and not damaged
   • Ensure barcodes are not too small or too large

Performance Tips

• Use high-resolution images for better OCR accuracy
• Ensure good lighting and contrast in images
• For batch processing, consider processing in smaller batches

Example Use Cases

• Drug Inventory: Scan medication barcodes and extract drug names, dosages, expiry dates
• Product Catalog: Process product images with barcodes and descriptions
• Document Processing: Extract both barcode and text information from documents

QR Code + Medicine ID Lookup

Setting Up Medicine Database

1. Prepare Your Excel File:

   • Create an Excel file (.xlsx) with medicine details
   • Each row should represent one medicine
   • Columns can include: Medicine Name, Manufacturer, Expiry Date, Batch Number, MRP, etc.
   • Place the Excel file in the project root directory (same folder as app.py)

2. Load Medicine Data:

   python load_medicine_data.py your_medicine_file.xlsx

   This will create a medicine_database.json file with unique IDs for each medicine.

3. QR Codes: Each medicine should have its own unique QR code that corresponds to its Medicine ID.

Using QR Code Lookup

1. Start the Web Application:

   python app.py

   Or use:

   python start_web_app.py

2. Access the Web Interface:

   • Open your browser and go to http://localhost:5000
   • Click on the "QR Code + Medicine ID Lookup" tab

3. Lookup Medicine:

   • Upload the QR code image for the medicine
   • Enter the corresponding Medicine ID (e.g., HM002, ABC123, etc.)
   • Click "Lookup Medicine Details"
   • The system will validate the QR code and display medicine details if both QR and ID match

Important Notes

• QR Code Validation: Each medicine has a unique QR code that must be uploaded
• Medicine ID: Each medicine has a unique ID from your Excel file (e.g., HM002, ABC123)
• Security: The system requires both the correct QR code AND the correct Medicine ID to work
• Matching Required: The QR code and Medicine ID must match each other for successful lookup

File Structure

project/
├── images/                    # Input images folder
├── uploads/                   # Temporary upload folder (auto-created)
├── templates/                 # HTML templates for web app
│   ├── base.html
│   ├── index.html
│   ├── results.html
│   ├── medicine_result.html
│   └── database.html
├── barcode_scanner.py         # Main scanner script
├── app.py                     # Flask web application
├── load_medicine_data.py      # Load medicine data from Excel
├── setup_and_run.py          # Automated setup script
├── requirements.txt          # Python dependencies
├── README.md                 # This file
├── barcode_text_results.json # Output results (generated)
├── medicine_database.json    # Medicine database (generated from Excel)
└── your_medicine_file.xlsx   # YOUR EXCEL FILE - Place it here!