Vision-Based Structural Adhesive Detection for Electronic Components on PCBs

Structural adhesives or fixing glues are typically applied to larger components on printed circuit boards (PCBs) to increase mechanical stability and minimize damage from vibration. Existing work tends to focus on component placement verification and solder joint analysis, etc. However, the detection of structural adhesives remains largely unexplored. This paper proposes a vision-based method for detecting structural adhesive defects on PCBs. The method uses HSV color segmentation to extract PCB regions, followed by Hough-transform-based morphological analysis to identify board features. The perspective transformation then extracts and rectifies the adhesive regions, and constructs an adhesive region template by detecting the standard adhesive area ratio in its corresponding adhesive region. Finally, template matching is used to detect the structural adhesives. The experimental results show that this approach can accurately detect the adhesive state of PCBs and identify the qualified/unqualified locations, providing an effective vision-based detection scheme for PCB manufacturing. The main contributions of this paper are as follows: (1) A vision-based structural adhesive detection method is proposed, and its detailed algorithm is presented. (2) The developed system includes a user-friendly visualization interface, streamlining the inspection workflow. (3) Actual experiments are performed to evaluate this study, and the results validate its effectiveness.