In the context of increasing automation in bottling lines, the study by Asif Hussain and colleagues proposes a low-cost computer vision system for real-time cap inspection, aiming to improve quality control by reducing reliance on expensive solutions and manual checks.

The research is based on a Raspberry Pi platform integrated with image processing techniques and open-source computer vision algorithms. Specifically, the system acquires images using a dedicated camera and analyzes them using tools such as Haar Cascade, SIFT, and the OpenCV library, identifying defects such as missing, misplaced, or damaged caps directly on the conveyor belt.

A distinctive feature is the low-cost and modular approach, which allows for reliable performance without the need for advanced industrial sensors.

The system is capable of real-time operation, immediately reporting anomalies via visual notifications (LEDs) and remote monitoring interfaces, improving the responsiveness of the production line.

The tests, conducted on moving bottles, demonstrate good accuracy in detecting the presence and correct positioning of the cap, confirming the validity of the approach even under real-world operating conditions.

In summary, the study highlights how in-line quality is no longer just a matter of advanced technology, but of intelligent and accessible integration, opening up new digitalization opportunities even for plants with limited resources.

Bibliographical references: Hussain, S.A., Khan, M., Babu, J.C., Hussain, S.J. and Hu, Y.-C. (2026). An Automated System to Identify and Detect the Faults in Bottle Cap Production and Visual Inspection Using Raspberry Pi. In Digital Twins and ESG (eds S. Mondal, A. Kumar and M. Khan). https://doi.org/10.1002/9781394303243.ch10