Glass Jar Neck Inspection
Detect defects in the threaded necks of glass containers
Graphical programming environment for deep learning-based industrial image analysis
Glass jars and other containers for food products with screw tops can suffer many kinds of impact damage to the threaded neck. A wide range of chips, cracks, inclusions, and other defects can indicate the possibility of a glass chip within the jar, a potential poor seal when the lid is installed, or danger to the final consumer. Each jar must be inspected from above for defects before proceeding to filling and sealing. Even small defects in a food container can lead to consumer dissatisfaction, particularly since glass packaging is often used for more expensive luxury goods sold in smaller volumes.
The variety of types and locations of possible damage, along with the transparency, reflectivity, and luminance variation of the glass make it almost impossible for conventional machine vision to reliably identify defects while passing acceptable jar necks.
Cognex AI-enabled solutions solve the problem of detecting subtle defects in threaded glass necks. They train on a set of images of acceptable glass container necks. The defect detection tool then identifies anomalies such as chips, inclusions, and cracks, while accepting the wide range of possible appearances of the glass neck, disregarding glints, hot spots, and refractions.
Jars and other glass containers reach the consumer with tight seals and with a much-reduced risk of glass chips or other dangerous physical contaminants.