Automating Quality Control for a Leading Pizza Chain
How Algovectra developed a state-of-the-art computer vision system to ensure every pizza leaving the Vosíme kitchens meets their exacting standards.
The Challenge
In a high-throughput kitchen environment, maintaining consistent quality across hundreds of orders a day is a monumental task. The client, Vosíme, needed a way to verify that every pizza had the correct ingredients, proper distribution, and no unwanted burn marks before it reached the customer. Manual inspection was prone to human error during peak rushes, leading to inconsistent customer experiences and increased refund claims.
The Solution
Algovectra designed an end-to-end edge AI system known as the Pizza Checker. We deployed custom-built camera rigs housing Raspberry Pi compute modules equipped with high-quality Sony IMX477 sensors directly over the assembly lines.
- Edge ProcessingImages are captured and processed locally using highly optimized OpenCV models, ensuring immediate feedback without relying on cloud latency.
- FastAPI BackendA robust Python backend handles incoming data streams, evaluates the pizza criteria against the POS order, and logs metrics.
- Kitchen Tablet IntegrationStaff receive real-time visual alerts (pass/fail) directly on their workstations, preventing flawed items from being boxed.
"The Pizza Checker system fundamentally transformed our quality assurance. It acts as an untiring supervisor, allowing our kitchen staff to focus on preparation rather than second-guessing every order. It's an indispensable part of our infrastructure now."