Here is the abstract you requested from the imaps_2014 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Automatic Segmentation Method for Segmenting Chip Package and PWB Regions during Warpage Measurement of Unpainted PWB Assembly|
|Keywords: Fringe Projection, Warpage Measurement, Segmentation|
|Among the various warpage measurement techniques, the fringe projection techniques (laser fringe projection (LFP) and digital fringe projection (DFP)) are recent trends because of their noncontact, full-field, and high-resolution measurement capabilities for measuring the warpage of chip packages, printed wiring boards (PWBs) and PWB assemblies (PWBAs). When using the fringe projection techniques, reflective painting is generally sprayed on the sample surface to ensure uniform surface reflectance and to obtain better image contrast for the measurement process. However, because painted samples may no longer be re-used, and the spray-painting process is not suitable inline, warpage measurement of unpainted chip packages, PWBs, and PWBAs using the fringe projection techniques is required. One of the important tasks for measuring the unpainted PWBAs is segmentation of the chip package and PWB regions in an unpainted PWBA images so that the separate warpage of the chip package(s) and the PWB in the PWBA can be determined. In this paper, an automatic image segmentation algorithm that is used to segment the chip package and PWB regions in an unpainted PWBA image will be presented. Because the patterns such as the copper patterns and traces and inscriptions on a PWBA can cause measurement error, the patterns and inscriptions are automatically masked out during the segmentation. The experimental results will show that the algorithm can be used for various PWBA samples.|
|Sungbum Kang, Ph.D. Student
Georgia Institute of Technology