ExpireEye provides barcode and QR code scanning functionality using OpenCV’s computer vision library to quickly identify and add products to your inventory.
QR code scanning
The API includes a dedicated endpoint for scanning QR codes from uploaded images:
How it works
Upload image
You send an image file containing a QR code to the endpoint:@app.post("/qr", tags=["QR Code"])
async def decode_qr_code(request: Request, file: UploadFile = File(...)):
access_token = request.state.user
user_id = access_token.get("userId")
Decode image with OpenCV
The image is converted from bytes to a numpy array and decoded:image_bytes = await file.read()
np_arr = np.frombuffer(image_bytes, np.uint8)
# Decode image using OpenCV
img = cv2.imdecode(np_arr, cv2.IMREAD_COLOR)
Save uploaded image
The uploaded file is saved to the uploads directory:filename = (
str(file.filename)
if file.filename
else f"qr_{datetime.now().strftime('%Y%m%d_%H%M%S')}.jpg"
)
file_path = os.path.join("uploads", filename)
with open(file_path, "wb") as buffer:
shutil.copyfileobj(file.file, buffer)
Detect and decode QR code
OpenCV’s QRCodeDetector is used to detect and decode the QR code:# Initialize QRCode detector
detector = cv2.QRCodeDetector()
# Detect and decode
data, bbox, _ = detector.detectAndDecode(img)
if bbox is not None and data:
return {"message": "QR Scanned successfully", "data": data}
else:
return {"error": "No QR code found in the image"}
curl -X POST "http://localhost:8000/api/qr" \
-H "Authorization: Bearer YOUR_TOKEN" \
-F "file=@/path/to/qr-code.jpg"
{
"message": "QR Scanned successfully",
"data": "Product information from QR code"
}
Barcode lookup system
ExpireEye includes a barcode-to-product mapping system for quick product identification:
Get all barcodes
GET /api/product/barcodes
@router.get("/barcodes")
def get_products_barcode(db: Session = Depends(get_db)):
products = db.query(Product).all()
result = [
{
"productName": product.name,
"barcode": product.barcode,
}
for product in products
]
return {"products": result}
Barcode generation
When you add a product to the inventory, a barcode is automatically generated:
def generate_product_barcode(product_name: str) -> str:
"""
Generates a simple barcode for a product based on its name.
Uses a predefined mapping for known products. For test purposes.
"""
product_name = product_name.title()
return product_barcode_map.get(product_name, "UNKNOWN-20")
Predefined barcode mapping
The system includes barcodes for common products:
product_barcode_map = {
"Apple": "AP29483",
"Banana": "BN10375",
"Lays": "8901491101813",
"Kurkure": "KK42196",
"Tomato": "TM77238",
"Potato": "PT66120",
"Onion": "ON95742",
"Mango": "MG33421",
"Carrot": "CR16027",
"Cucumber": "CU48290",
"Milk": "ML12345",
"Bread": "BR67890",
"Eggs": "EG54321",
"Rice": "RC98765",
# ... additional mappings
}
Reverse barcode lookup
You can look up a product by its barcode:
def get_product_name_from_barcode(barcode: str, db: Session = None) -> str:
"""
Retrieves the product name from a given barcode.
"""
product = db.query(Product).filter(Product.barcode == barcode).first()
if product:
return product.name
else:
return None
Image processing with OpenCV
The system uses OpenCV for robust image processing:
import cv2
import numpy as np
# Convert image bytes to numpy array
np_arr = np.frombuffer(image_bytes, np.uint8)
# Decode image
img = cv2.imdecode(np_arr, cv2.IMREAD_COLOR)
# Initialize QR detector
detector = cv2.QRCodeDetector()
# Detect and decode
data, bbox, _ = detector.detectAndDecode(img)
OpenCV’s QRCodeDetector can handle various QR code formats and orientations automatically.
- JPEG (.jpg, .jpeg)
- PNG (.png)
- BMP (.bmp)
- TIFF (.tiff)
All uploaded images are temporarily stored in the uploads/ directory for processing.
Integration with product detection
QR code scanning can be combined with AI-based product detection for a complete inventory management workflow:
- Scan QR code to identify product
- Use YOLO detection to verify product visually
- Automatically add to user inventory
- Track expiry dates and receive notifications
Make sure the uploads/ directory exists and has proper write permissions before using the QR scanning feature.
Error handling
The system handles common scanning errors:
- No QR code detected: Returns error message with guidance
- Invalid image format: OpenCV will fail to decode the image
- Corrupted file: Returns appropriate error response
if bbox is not None and data:
return {"message": "QR Scanned successfully", "data": data}
else:
return {"error": "No QR code found in the image"}