Automatic detection of stamping defects in leadframes using machine vision: Overcoming translational and rotational misalignment.

This paper presents an algorithm for the automatic inspection of leadframes to detect stamping defects in the presence of translational and rotational misalignment. The algorithm uses several image processing operations such as blob analysis, morphological closing and image subtraction to detect the stamping defects. The algorithm has been successfully tested on leadframes having simulated and real defects on multiple modules. The proposed algorithm overcomes both translational and rotational misalignment of the leadframe, thus eliminating the need to synchronize the leadframe movement and camera image capture. The system can be integrated in the manufacturing line for inspecting both continuous reel and individual cut leadframes. [ABSTRACT FROM AUTHOR]