High volume producers of 2-D parts like electronics components have long used vision systems to inspect their parts. However, mainstream manufacturers of machined components are just beginning to adopt this technology.

Ring Light – For top lit edge measurements, ring light is generally best. Source: Wilcox Associates.

Today’s vision and multisensor systems, with 3-D, Computer Aided Design (CAD)-based programming and operating software provide users with a new kind of CMM capable of measuring critical features that cannot be measured with conventional touch probes because they are either too small, inaccessible or subject to deformation on contact. They also offer significant speed advantages, since a vision system can measure multiple features simultaneously when they are in the same frame.

Some potential users are reluctant to try this technology because of perceived difficulties in using it for controlling measurement parameters like part illumination and contrast. These might have been valid reservations five years ago but today developers of vision measurement software offer adaptive products that allow even inexperienced users to make fast and accurate adjustments to critical illumination parameters. These result in consistent measurements from part to part and vision machine to vision machine.

Coaxial Illumination – Coaxial is the best choice for most surface measurements. Source: Wilcox Associates.

Illumination Types

A vision system will only measure what it can see, so one must light the part in some fashion. The common choices are sub-stage (backlight), coaxial (top), and ring-light. Sub-stage illumination is the method of choice for measuring parts in profile. In this case, contrast is sharp and illumination adjustments straightforward.

Coaxial is the best choice for most surface measurements. For top lit edge measurements, ring light is generally best. This light is projected at the same focal point as the camera, but at a different angle of incidence to heighten contrast and make edges more easily detectable.

Sub-stage Illumination - Sub stage illumination is the method of choice for measuring parts in profile. In this case, contrast is sharp and illumination adjustments are straightforward. Source: Wilcox Associates.

Edge Detection

Now here is what is a little scary to some people. Finding the real edge can be tricky. Unlike tactile probes, cameras do not touch the edge they are seeking to measure. So, edge detection must rely on the accurate interpretation of data the vision software receives from the camera. Fortunately, advanced vision measurement software is able to tune its edge detection algorithms to account for both the part’s surface and the illumination conditions allowing it to find each feature’s edge accurately.

Generally, software will use a dominant edge algorithm to select the edge of a part when using sub-stage illumination. Measuring top-lit parts with a high surface finish is more problematical. In this case users may choose a specific edge algorithm, which allows the detection of the feature in question based not only on contrast but its shape and location.

If there are grind marks on the part, which might confuse the camera when top lighting is used, the software can apply another type of algorithm that chooses the most dominant edge out of possible candidates within the camera's field of view. Software producers have spent enormous amounts of time developing these proprietary algorithms, so the user should be aware that there are differences.

Ideal Contrast

If one puts his or her finger in a fast moving stream of water, the water will bend slightly around the finger to contact areas of flesh that are out of the stream's direct line. Light does the same thing. If one puts a precision roller on a vision system and measures it, he or she will get one value. If one turns the lighting intensity higher and measure again, he or she is likely to get a somewhat smaller value. This is because higher intensity light has a tendency to bend around cylindrical part and the camera interprets the distance between edges as shorter than the actual diameter.

Adjusting the light source to deliver an optimal contrast level easily eliminates this and other potential problems associated with lighting intensity. However, care is necessary. Too much light washes out the pixels and too little obscures them. Until recently, users had to make these adjustments manually and that subjectivity was a justifiable rap against vision measurement. Today, however, special algorithms are available with some vision systems that optimize contrast levels automatically. Literally, at the touch of a button, the algorithms will make rapid iterative adjustments until they find an optimal and consistent contrast level.

Consistent Intensity

Of course, lighting intensity varies according to the type of light source used and the ambient lighting conditions. Halogen bulbs blast out a lot of light and are used when more intense lighting is required. Unfortunately, they consume more energy and produce a lot more heat. Whenever it’s possible, LED bulbs are the illumination source of choice. They both reduce the cost of energy and eliminate the need to dissipate heat form the part.

Both halogen and LED bulbs have a certain amount of play in their intensity as power ramps up or down for different levels of lighting. In general, the intensity of halogen bulbs varies directly with the increments of adjustment up to about 75% of the maximum energy available. This is the point where the bulb is incapable producing any more light. LEDs will produce accurate incremental lighting intensity changes over the first 90% of the power range.

Avoiding both these sources of lighting inconsistency and those created when moving parts to different vision systems or changing ambient lighting is a simple matter of calibrating the machine’s illumination. Advanced vision software products now have algorithms that let users do this quickly and automatically making it no more difficult or esoteric than calibrating a probe on a CMM.

None of these issues should cause manufacturers who could benefit from the many advantages of vision measurement to shy away from it. Vision equipment with CAD-based programming and operating systems are really just another type of CMM. As is the case with the best tactile measurement software, some vision software now incorporates sophisticated, assistive technologies that allow the user to measure parts while the software automatically handles the technical details automatically. If you can use the CMM, you should have no problem using a vision or multisensor measurement system.