Anomaly detection is the most advanced application of Edge-Vision-4.0. We go beyond merely locating faults: using Deep Learning and trained neural networks, we classify every defect immediately. This enables you to address the root cause of the defectsupstream in your process and minimize waste.
Based on a dataset of images pre-classified by the customer, we use artificial intelligence to build a powerful neural network. This model is then deployed for in-line quality control, allowing the system to continuously trace complex, unknown, and hard-to-see defects in materials like textiles, plastics, and battery film.
Spot Detection
Identifies spots in extruded films and sheets.
Stripe Detection
Detects length-direction stripes from die contamination in extruded sheets.
Foreign Object Detection
Finds foreign objects in non-woven materials
Anomaly Detection
Identifies holes, stains, stitching errors, and printing anomalies in textiles
Defect Detection
Detects fractures and defects in battery film coatings
Color and Glossiness Detection
Quantifies color variations and identifies surface glossiness variations or transparency issues.
The speed and precision of our anomaly detection are guaranteed by the integration of advanced technologies:
1
4K Camera chip technology
Enables the detection of anomalies down to hundredths of a millimeter.
2
Edge computing
The 4K image data is processed locally in real-time, avoiding data bottlenecks and cloud latency. This ensures immediate reaction to every detected defect.
3
Deep learning & data compression
Translates large volumes of defect data into immediate, actionable metrics and decisions for the operator.

Our technology proves its worth daily in the most demanding production environments, tackling critical quality challenges.
Federal Eco Foam, a leading foam manufacturer, shares how Edge-Vision-4.0's AI inspection capability helped them eliminate critical contaminants (like hidden plastic or metal) in their complex recycled materials, transforming their quality control and significantly reducing waste.