|
|
Related to applications
like circuit processing is the need to inspect the processed part.
Vision systems to perform this inspection immediately after processing
can reduce the financial impact of adding further value to a defective
part, in addition to screening the production run to improve the
quality of shipped product. Laser processes can sometimes be used to
correct the defect (e.g. remove a short from a patterned ITO film) or
mark the part for downstream rejection without additional value added
to the part, and so improve yield and profitability. |
 |
|
The cameras employed can be either be area scan (normal 2-D) or
line-scan cameras. The latter are designed to capture stripes of
information from rapidly-moving parts, and this information is used to
reconstruct an image of the part. With special design, high
sensitivities and high speed can be obtained simultaneously.
Built-in calibration targets allow for rapid, automated system
self-calibration, minimizing the requirement for operator or
maintenance intervention. |
 |
|
Vision processor boards are typically used to collect, reconstruct,
and analyse the camera output, leaving the system computer free to
perform its main functions of sub-system coordination and user
interface. Analysis can take the form of point location, finding
boundaries (transitions), or performing blob analysis on complex
shapes of any size within the camera field of view. Product can
be assigned a grade, or a simple Pass/Fail decision applied. In
either case all parameters are logged and statistical output is
available as required. |
|
The illumination source is the other important piece of hardware.
This must be selected in combination with the camera sensitivity and
required illumination uniformity. LED sources are frequently
used since they are now readily available in a variety of wavelengths,
and have the benefit of low power consumption. |
 |
|
