Capacitor Soldering Inspection
Identify small defects in the solder, which can cause wiring breakages, shorts, and other electrical problems.
Graphical programming environment for deep learning-based industrial image analysis
For a component like a mouse diode to be mounted without interference to its electrical connection, solder resist must be applied cleanly to a bare board. Even small defects in the solder can cause wiring breakages, shorts, and other electrical problems. These defects vary in size, shape, and appearance due to specular glare. It is difficult to program an automated inspection that tolerates significant part variation under these conditions.
Cognex Deep Learning quickly identifies the solder resist on a diode when other methods struggle to inspect under the same lighting conditions. The assembly verification and part location tool trains on a set of representative images of solder resist to learn the normal appearance of “good” and “bad” solder. During runtime, the tool fixtures and locates resist on the PCB, despite variations in specular glare. During the second stage of the inspection, the solder resist must be inspected to find any functional anomalies, such as bridging, peaking, or gapping. Using the defect detection tool in supervised mode, the user can train the tool on a representative set of known “good” solders and “bad” solders with labeled defects.
Based on these images, Cognex Deep Learning learns the natural texture of the mouse diode, as well as the normal appearance of its solder. Additional images can be added to the training set during validation testing to reflect additional examples and optimize the model. Various parameters can be adjusted during the training and validation phase to help account for variations in appearance to correctly detect all the diodes with defective solder.