Impact

95%

Bullseye grading agreement, well above the 75% target

<0.20"

Measurement error against a 0.50-inch threshold

50ms

Real-time processing latency, with one model scaling across multiple packaging lines

Enhancing packaging quality control with computer vision

Challenge

A leading beverage bottler needed a scalable, real-time way to quality-control shrink-wrapped bottle packages, with particular focus on bullseye handles and film creases that affect package strength. Existing proprietary camera systems required per-line calibration, missed defects, and introduced quality variability that slowed operations.

The company was looking for a faster, more accurate, and scalable way to automate shrink-wrap inspection across multiple production lines, with reduced operator subjectivity and lower deployment cost per plant.

Solution

OneSix designed and built an edge-based, AI-driven computer vision system that detects bullseye defects and packaging creases using a single multi-task model. The system converts pixel data into real-world measurements and applies automated PASS / WARN / FAIL grading, removing operator subjectivity from the inspection process.

The solution runs on affordable edge devices (Jetson Orin NX), supporting four to five cameras per line with minimal hardware and 50ms processing latency. A single trained model scales across multiple packaging lines with minimal per-line tuning, lowering deployment cost and accelerating rollout across plants.

Integration with SCADA systems and Azure provides real-time monitoring on the floor and historical data capture for continuous improvement. In production, the system achieved 95% bullseye grading agreement against a 75% target and measurement error between 0.05 and 0.19 inches against a 0.50-inch threshold.