Machine Vision
Anomaly Detection
Anomaly detection trains the deep learning model only on images of good products; the model flags every product that departs from the normal appearance.
Defect classification needs many examples of every defect type, yet faulty products are rare on a production line and a new kind of defect can appear any day. In anomaly detection the model learns what a good product looks like and gives every new image an anomaly score; a product scoring above the threshold is rejected as NOK.
The heat map shows which area of the product departs from normal. Defects such as dented or torn packaging, missing labels, stains or wrong placement are shown to the operator visually and recorded.
The example in the image comes from a demo we prepared for packaging inspection: on the left a good box is rated "good" with a low score, on the right a box with a defect at its bottom is flagged as an "anomaly" with a high score.
Highlights
- Trained on good products only
- Catches unseen defect types
- Defect location with a heat map
- Adjustable score threshold
Application areas
- Packaging & labels
- Food
- Textile
- Castings & metal surfaces
