Cosmetic defect inspection
Capture defects on challenging packaging surfaces
Graphical programming environment for deep learning-based industrial image analysis
One of the biggest challenges when dealing with cosmetic, or surface, defects on consumer packaged goods is that they are dynamic, often caused by the forming process. Typical defects like hits, scratches, or stains may be indiscernible on a part’s textured surface during early production. These defects only become visible under specific lighting conditions later in the production process. While the cost of late detection can be painfully high, so can false rejects. This inspection is especially important in regulated industries where poor packaging quality can lead to recall events or customer complaints.
Conventional vision technology can often miss complex cosmetic packaging defects such as bubbles in labels, color degradation, scratches, cracks, overprint, and over-wrap or under-wrap issues. These types of unpredictable defects or variations are easy to discern by human inspectors, but very difficult to program with rule-based machine vision algorithms.
AI-enabled image analysis software detects cosmetic defects on rough and textured metal surfaces as reliably as human inspectors, but with the speed of a computerized system. The defect detection tool catches defects on coarse material with standard illumination, even when image quality is poor, by forming a reliable model of the part’s shape and texture based on training images. From here, it identifies deviations in the surface texture as anomalies and uses a classification tool to classify them as hits or scratches.