Printed Circuit Board Assembly Verification
Identify and count PCB components to ensure correct assembly using deep learning software
During final assembly verification, 2D and 3D machine vision systems traditionally inspect PCBs for the presence and correct placement of LEDs, microprocessors, and other surface mount devices. Wrongly positioned or missing components can impact a PCB’s performance and lifespan. Manufacturers must be vigilant for scratched, twisted, bent, or missing pins. A chip has such low tolerances for error that any flaw, even the most superficial, is cause for rejection.
These errors must be caught before PCBs are assembled into devices or shipped to customers. Yet slight variations in appearance—whether due to subtle lighting contrasts, changes in perspective and orientation, or glare on metallic surfaces—can confuse an automated inspection system. Parts which are close together are difficult for a machine vision system to distinguish as independent components. Programming these inspections into a rules-based algorithm is time-consuming, prone to error, and challenging for a field engineer to maintain. Human inspectors, though capable of identifying these components, cannot meet the high-speed processing demands.
Cognex ViDi deep learning image analysis software offers a field-maintainable solution for PCB assembly inspection. The ViDi Blue-Locate tool learns to identify the components from images labeled with locations of each part type, building a reference model of each component within a single tool. This tool generalizes the distinguishing features of the parts based on their size, shape, and surface features and learns their normal appearance, as well as their general location on the board. The system is also optimized to work with images that are low-contrast or are poorly captured.
In production, ViDi Blue-Locate analyzes all the relevant areas of the board to locate and identify each component, despite changes in appearance. This solution is able to determine whether the components are present or absent and verifies that the correct component is being used. In this way, the tool can provide a reliable solution to automate complex PCB assembly inspection.