Automatic Defect Detection System For Leadframe Inspection
Loading...
Date
2004-04
Authors
Rajamony, Bhuvanesh Abhinesh
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Defect detection and classification are important for both product quality assurance and
process improvement in the maimfacturing industry. Machine vision systems offer several
beneficial features such as consistency, accuracy and round the clock repeatability. This
presents the results of the development and implementation of such a machine
system to automate the inspection of leadframes. Inspection of stamping defects and
defects are the focus of this research. Stamping defects are caused due to
improper stamping which primarily comprise of irregularities in meeting geometrical
specifications and material removal. Whereas slug mark defect is caused by the stamping
tool leaving an undesirable indentation on the leadframe after stamping. In this respect,
several algorithms have been developed to inspect improper stamping not only for the
critical internal leads but also the entire leadframe including the outer railing which
contains the pilot holes. The proposed method follows three steps to evaluate the quality of
the product. The first step consists of capturing images of the leadframe to be inspected.
Next, using mathematical morphological processing, the image is subtracted from a defect
free template image. The last step is to mark the location and display the defects. The
proposed algorithms were tested cin a variety of leadframes, and the experimental results
are presented. The algorithms were extended and also tested for inspection of Ieadframes
in the presence of translational and rotational misalignment. Experimental results showed .
that the proposed algorithms can be employed not only for inspecting individual cut
leadframes but also for continuous inspection in the manufacturing line. It has been
possible to detect and display defects in a fast and efficient way with minimal errors.
xiii
Description
Keywords
Quality assurance