Skip to main content

How to Decode Barcodes and Industrial 2D Codes with Python

Β· 6 min read
Serhii Hrekov
Serhii Hrekov
software engineer, creator, artist, programmer, projects founder

Building a code generator is only half the battle. In this walkthrough, we will build a high-performance scanning system capable of identifying and decoding multiple barcodes and 2D codes (QR, Data Matrix, etc.) from both static images and real-time video streams.

1. The Power Couple: OpenCV and PyZbar​

To read barcodes effectively, we use OpenCV for image manipulation and PyZbar for the heavy lifting of decoding.

  • OpenCV: Handles the "eyes" of our application-opening the camera, converting colors, and drawing feedback on the screen [4.2].
  • PyZbar: A Python wrapper for the ZBar library, which is a world-class engine for scanning barcodes across many formats [4.1].

πŸ’» Installation​

On most systems, you will need to install the C-library for ZBar before the Python wrapper:

  • Windows: Automatically included with the pip package.
  • macOS: brew install zbar
  • Linux: sudo apt-get install libzbar0

Then, install the Python libraries:

pip install opencv-python pyzbar

2. The Scanning Pipeline​

A robust scanner doesn't just "look" at an image; it processes it. The pipeline looks like this:

  1. Pre-processing: Convert to grayscale and apply a slight blur to reduce noise [4.3].
  2. Decoding: PyZbar scans the processed image for known patterns.
  3. Localization: We extract the coordinates of the code to draw a bounding box.
  4. Parsing: The raw bytes are converted into a readable string.

3. Static Image Scanner (Code Walkthrough)​

This script will take a single image file and print every barcode it finds inside it.

import cv2
from pyzbar import pyzbar

def scan_image(image_path):
    # 1. Load the image
    image = cv2.imread(image_path)
    
    # 2. Pre-process: Convert to grayscale
    # This makes the black-and-white patterns easier to distinguish
    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
    
    # 3. Decode all barcodes in the image
    barcodes = pyzbar.decode(gray)
    
    for obj in barcodes:
        # Extract the data as a string
        data = obj.data.decode("utf-8")
        type = obj.type
        
        # 4. Localization: Get coordinates for a bounding box
        (x, y, w, h) = obj.rect
        cv2.rectangle(image, (x, y), (x + w, y + h), (0, 255, 0), 2)
        
        print(f"Found {type} Code: {data}")
        
    # Show the result with the bounding box
    cv2.imshow("Scanner Result", image)
    cv2.waitKey(0)

# scan_image("my_labels.png")

4. Real-Time Webcam Scanner​

For industrial or retail apps, you need a live feed. This script initializes your webcam and decodes data at 30+ frames per second.

import cv2
from pyzbar import pyzbar

def live_scanner():
    # Initialize the camera (0 is usually the default webcam)
    cap = cv2.VideoCapture(0)
    
    print("Scanner started. Press 'q' to quit.")

    while True:
        ret, frame = cap.read()
        if not ret:
            break
            
        # Optional: Pre-processing for speed
        # gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
        
        # Detect and decode
        decoded_objects = pyzbar.decode(frame)
        
        for obj in decoded_objects:
            # Draw green box around detected code
            points = obj.polygon
            if len(points) > 4: 
                hull = cv2.convexHull(np.array([point for point in points], dtype=np.float32))
                hull = list(map(tuple, np.squeeze(hull)))
            else: 
                hull = points
                
            # Draw the lines
            n = len(hull)
            for j in range(0, n):
                cv2.line(frame, hull[j], hull[(j + 1) % n], (0, 255, 0), 3)

            # Overlay the decoded text
            data = obj.data.decode("utf-8")
            cv2.putText(frame, data, (obj.rect.x, obj.rect.y - 10), 
                        cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)
            
            print(f"Scanned: {data}")

        cv2.imshow("Live Scanner", frame)
        
        # Stop if 'q' is pressed
        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    cap.release()
    cv2.destroyAllWindows()

# live_scanner()

5. Multi-Format Handling​

One of the best features of pyzbar is that it handles multiple formats simultaneously. You don't need a separate "QR Scanner" and "EAN-13 Scanner." A single pyzbar.decode() call will return:

  • Standard Barcodes: EAN-13, UPC-A, Code 128, Code 39.
  • 2D Codes: QR Codes.

Note on Data Matrix/PDF417: While pyzbar is excellent for standard barcodes and QR codes, for specialized industrial codes like Data Matrix, you might need the pylibdmtx library.

πŸ“š Sources & Further Reading​