248 Unit Four Control Systems and Maintenance
Fundamentals of Machine Vision
Machine vision systems use video cameras and computers to translate
light energy into an image. These images are then used to determine part
orientation or another designated task. However, even the most effective
machine vision systems cannot achieve the level of perception and sophis-
tication of the human eye.
Some of the earliest machine vision systems were introduced in
the late 1970s. Early systems were expensive due to the high cost of the
computer memory needed to handle the images. They were also ineffi-
cient, because all computers, except mainframes, contained fairly slow
processors. Today, however, faster computers and expanded memory
capabilities make vision systems very practical. The data obtained can
be used to make decisions on the processing steps or other action taken
by either human operators or a robot. Vision systems can also be used in
conjunction with material inventory and material flow within a factory.
Four functions take place during image processing—acquisition,
preprocessing, analysis, and interpretation, Figure 10-9.
Figure 10-8. The video camera “eyes” of this system are suspended over the conveyor to scan parts
as they travel along the conveyor. (FANUC Robotics)