High-Speed OCR & Barcode Verification
Problem Statement: In the mass production of product labels, ensuring that the printed human-readable text (OCR) matches the machine-readable barcode is critical. Manual inspection on high-speed reel-to-reel machines is impossible, leading to risks of mixed data, missing prints, or unreadable codes reaching the end customer.
The Challenge: The client needed a system capable of inspecting every single label on a moving roll. The challenge involved not just reading the data, but performing a Dual Validation Cross-check—verifying that the alphanumeric string (OCR) perfectly matches the decoded Barcode data in real-time, while also controlling the machine’s motor to stop immediately upon detecting a defect.
Our Solution: LabVIEW Vision Architecture
We engineered a robust Machine Vision solution integrated directly into the Reel-to-Reel machinery.
The system utilizes LabVIEW with the NI Vision Development Module to process images captured by a high-resolution Basler Camera (GigE/USB3). By implementing a specialized Backlight Unit, the system detects label opacity and shadows to trigger precise image acquisition.
Key Technologies Implemented
| Technology | Role in System |
|---|---|
| LabVIEW & NI-IMAQdx | Core image processing engine for OCR and Barcode decoding. |
| Basler acA1920-25gm | High-speed industrial camera for capturing clear images on moving conveyors. |
| Dual Validation Algorithm | Logic that compares OCR results against Barcode content to prevent data mismatch. |
| Industrial I/O Integration | Direct hardware interface to control the feed motor (Start/Stop/Emergency) based on inspection results. |
Business Results
- 100% Automated QC: Eliminated manual sampling; every label is inspected for readability and accuracy.
- Production Control: Integrated order management allows operators to set target amounts, with the system auto-stopping when the order is complete.
- Data Integrity: Prevented “Mixing/Wrong Data” errors by cross-checking OCR numbers against Barcodes.
Expert Insight: For transparent or semi-transparent labels, standard front-lighting often fails. We utilized a Backlight technique to create a silhouette effect, ensuring the camera triggers at the exact edge of the label regardless of the print design.
Looking to automate your quality control? Contact us at: wisit.paewkratok@gmail.com