Razor Blade Defect Detection
Inspect complex mesh foils for subtle cosmetic defects
The cutting blades of an electric razor oscillate underneath a perforated foil. The foil’s many holes trap the beard hair so it can be cut and are also sharp on the underside to act as secondary blades. The pattern of holes on the foil is complex, to catch a wide range of hair types, and can vary from one part of the foil to another. The foil material is highly reflective.
The difference between an acceptable and an unacceptable mesh foil is subtle, but any cosmetic defects in a premium consumer good can result in lost sales and brand degradation. The functionality of the razor can also be impaired by any larger misalignment in the foil.
The complexity of the foil surface as well as its reflectivity make it difficult for conventional machine vision to distinguish between defects and acceptable variations when visually inspecting mesh foils before installation on the razor.
Cognex Deep Learning’s defect detection tool trains on a set of images of acceptable mesh foils allowing it to easily identify anomalies that mark the foil as outside of acceptable cosmetic limits.
If foil design or pattern changes, the defect detection tool can be retrained with a set of images of the new foil type and placed back online, inspecting, and detecting anomalies in the new foils without the need for reprogramming.