Computer Vision for Automated Quality Control in Industrial Bakeries

Overview

A large German food company operating was using computer vision quality control to monitor products on its industrial bakery lines moving through production lines. The system already included trained detection models, but the overall pipeline needed to become more modular, easier to extend, and more reliable for product tracking and counting. 

Challenge: Computer Vision Quality Control at Line Speed

Industrial bakery production is fast-moving and visually complex. A single camera image can show many bread rolls, loaves, or other bakery products at the same time. These products often move close together, partially overlap, or leave and re-enter the camera’s view. Detecting them in one image is only the first step towards counting them reliably.

The system also needs to recognise whether a product in the current frame appeared in previous frames. Otherwise, it could count the same bread roll twice, or miss it entirely.

Our Solution: Modular Computer Vision Quality Control

iits-consulting refactored parts of the customer’s existing computer vision framework into a more modular structure. The customer team can now build, replace, and extend individual processing modules within their AI workflow. 

A central part of our work was improving the existing product-tracking logic. The system tracks each product detected by existing models across multiple video frames. To count products consistently, an invisible counting line was defined in the camera image. When a tracked product crossed this line, it was counted once. 

The tracking pipeline uses a Kalman filter, which the team reworked and tuned to better handle the movement of many bakery products through the image. The existing recognition models stayed unchanged. The work focused on improving the structure, maintainability, and reliability of the existing tracking and counting workflow. 

Business Impact

The client now has a cleaner, more modular computer vision quality control pipeline for tracking and counting bakery products. The improved structure makes it easier for the team to adapt the framework, add new modules, and maintain the system with less effort. The tuned tracking logic also gives product counting and throughput monitoring a more stable foundation.

Wondering if AI could do the same for you? Start with a free AI Discovery Workshop.