Pre-Assembly Insertion Check and Battery Module Cosmetic Inspection
Check the battery’s integrity before the mobile phone undergoes final assembly
Graphical programming environment for deep learning-based industrial image analysis
During a pre-assembly insertion check, the contents of a phone are inspected for defects before the cover is assembled. Batteries can be damaged as they are guided onto the housing. Locating and inspecting the battery is difficult due to the confusing, busy background of the phone assembly. Deep learning vision software simplifies the automated detection and characterization of defects on the battery’s metal surfaces.
Cognex Deep Learning allows the manufacturer to check the battery’s integrity before the phone undergoes final assembly, and to differentiate between cosmetic and functional anomalies. Using the defect detection tool in supervised mode, an engineer can train the software on “good” images as well as “bad” images with labeled defects. From these images, the tool learns the battery’s normal appearance, including natural acceptable variations. Parameters can continually be adjusted during the training phase and validation period until the trained model correctly detects and segments all images with functional anomalies. Once deployed, the defect detection tool identifies and rejects batteries with defects.