OPET – Image Processing System for Label Inspection

In lubricant products, proper labeling is not merely aesthetic; it's critical for legal compliance, brand safety and customer communication. OPET wanted to ensure every product's label was positioned correctly, legible and complete. However, performing this inspection manually was slow and allowed faulty products to slip through. To meet this need, we developed an image processing system integrated into the production line that automatically detects labeling errors.

In the solution we developed, each product was analyzed with high-resolution cameras after labeling and checked instantly.

Core components of the system:

• Products were imaged with industrial cameras fixed on the production line
• Python-based image processing algorithms analyzed the label's alignment, position, orientation and integrity
• Mislabelled products were flagged by the system and separated within the production process
• A .NET interface provided operators with alerts and a control panel

What Did We Solve?

• We detected errors such as misplacement, incomplete printing or misalignment of labels
• The quality control process sped up and human error was minimized
• Brand integrity and regulatory compliance were maintained
• Provided capability for retrospective review and reporting

Results Achieved

• Labeling errors decreased by up to 90%
• Shipment of defective products was prevented, protecting brand image
• Manual inspection time was greatly reduced
• The quality process was digitized and became auditable