Understanding the technology of video metrology

Metrology is the science of measurement, and is a very broad topic - simply because so many things can be measured these days.

In manufacturing, the need to verify and validate dimensions of parts is crucial.

These dimensional measurements are done at many stages in the manufacturing process with a range of devices, from simple hand gauges to coordinate measuring machines (CMMs).

Often the choice of measuring device to be used is based on its resolution and range of the desired measurement.

One such technology is known as video measurement, performed with video measuring systems.

So, what is video metrology? A video measurement system is a machine that optically magnifies the image of a part captured by a camera and converted to a video signal that is analysed by electronics and software to determine edges and features.

The relationships among edges and features in one magnified image are monitored and tracked for comparison to other imaged areas to provide dimensions across the measurement range of the system's translation stages.

High-resolution imaging over large areas or volumes without any part contact makes video measurement a popular technology for manufacturers in many industries.

Video technology is also used for inspection.

The same benefits of magnifying a part to better resolve detail apply to devices used for visual inspection.

Such use isn't metrology, because the user subjectively analyses the image, and nothing is measured.

This use is more qualitative than quantitative.

There is no universal definition of 'video inspection', and some people refer to 'video inspection' when they mean 'video measurement'.

The capabilities described above can be provided in a number of ways.

The simplest devices have fixed magnification optics and require manual adjustment of focus and part positioning.

Such systems usually have small measurement ranges and fit on a bench top.

Automated systems have motorised zoom optics and stages, automatic focus, closed loop-positioning capability, and can run detailed, multi-step measurement routines - measuring sizes, dimensions, and relationships of single parts or of an array of parts - without any user intervention.

The operator loads the part, starts the measurement and the machine does the rest.

Automated systems are available in bench top and large-capacity, floor-mounted configurations.

The latest video measurement systems provide the resolution and accuracy necessary to perform critical dimensional measurements quickly.

Detailed numeric and graphical reports show specific results and flag out-of-tolerance conditions.

Engineers can compare measured results to computer-aided design (CAD) files and drawings for process control and monitoring.

What makes up a video measurement system?.

Optics - video measuring systems do not measure parts.

That's right: video measuring systems measure images of the part.

Therefore, an important aspect of an accurate video measuring system is good quality optics that faithfully submit images of the part to the camera.

Aberrations in low-quality optics might be interpreted as a measurement error in the part.

If the image of the part is distorted, that distorted image is what will be measured.

Zoom lenses allow measurement of part features across a range of sizes.

Motorised zoom lenses allow rapid, repeatable magnification changes.

Because measurement accuracy depends on knowing the relationship between part and image size, each change in magnification must be calibrated.

The best systems do this automatically.

For the image to be processed by software, it is converted to electrical signals by 'reading out' the signal levels from each of the pixels in a digital sensor - a camera.

Accurate autofocus and advanced edge-detection algorithms ensure accurate, repeatable measurement of every feature, at every magnification.

Lighting - because the image is so important, improving it with various lighting techniques is helpful.

Part features can be straight or curved, be on the perimeter or on the surface of the part, and be of different colours and textures; so a single type of illumination cannot satisfy every condition.

Back lighting - or profile illumination - is best for through-holes and the perimeter of the part.

On-axis top lighting illuminates the part from directly above, in line with the imaging optics.

Surface lighting at an oblique angle is helpful for highlighting subtle surface features.

For example, an illuminator made up of rings of light-emitting devices concentric to the optical axis allows illumination at selectable angles of incidence: the optics are illuminated as rings get closer to or farther from the part.

Segmenting the rings allows directional lighting by turning segments on or off.

This is especially helpful for highlighting edges lying in a particular orientation.

The best-designed ring lights allow illumination angle and direction to be changed without physically moving the light.

No matter the light source, its intensity must be adjustable to avoid overdriving the camera with too much light or having too little light for a sufficient signal-to-noise ratio.

Motion - Rarely does an entire part fit within the optical field of view.

This means that the part or lens must be moved until every feature to be measured is brought into the imaging field.

Take a flat part, for example.

Because this is a measuring machine, any motion in the X-Y plane of the part must be quantified.

Although the optical field of view might be a few millimetres across, X-Y stage travels of video machines extend to as much as one metre or more.

For accurate measurement over these distances, important parameters include straightness of travel, stage speed and positional resolution.

To measure a three-dimensional part, it is necessary to know the position of the image and the part throughout the measurement volume of the system.

This requires position sensing on macro and micro scales.

Take Z motion as an example: It is necessary to move the camera/optical assembly so the feature of interest is within the optical depth of focus at the particular magnification being used (depth of focus typically decreases as the magnification increases).

The entire camera/optical assembly must also retain its critical alignment as it moves.

Computer numerically controlled systems use motorised linear slides with scales to track position.

At the same time, on a micro scale, auto-focus takes over to maximise the image's sharpness.

System software notes the final position of the focused image relative to the system datum in all three axes.

Two categories of motion control are used.

Open-loop motion uses counts of a known increment and assumes a position is reached based on the stage drive motor moving a given number of multiples of that increment.

Closed-loop motion increases accuracy by adding a feedback device such as a linear scale to verify the stage position.

Total system accuracy depends on the type of motion system and the quality of the stage, motors, bearings, drive electronics and scales.

Mechanical structure - structural integrity of the system affects measurement accuracy and repeatability.

The XYZ axes must be orthogonal - true 90 degrees apart.

This requires a precision design and exacting assembly.

No part of the machine must move independently of any other because such an offset will affect measurements.

This requires a stable and damped mechanical structure, so video measuring systems are made of materials such as steel and granite for structural stability.

System integration - metrology software for automated video measuring systems allows an operator to create measurement routines by teaching the machine where to measure and which settings to use, or by creating the routines directly from the Cad files for the part.

For every measurement, the software repeats the settings for the XYZ position, zoom lens magnification, illuminator used and its settings, and tools used in the software itself for measuring within the image field.

Although a lot is going on, user interaction with the machine can be minimal.

Where we are today.

Video measurement technology has advanced dramatically over the past 30 years.

Computers are faster and cameras have better resolution and signal-to-noise ratios.

New software algorithms distinguish part features from background noise and artifacts.

Faster signal processing allows systems to measure and move the stages simultaneously as they follow an edge.

Stages are servo-driven.

System electronics with surface-mounted components on multi-layer boards offer increased speed, lower noise and less energy consumption.

Zoom lenses offer large magnification ranges, some with design advantages such as telecentricity.

Metrology software does much more than before and still is simply 'point and click'.

Selecting a system.

With so many variations in parts that make up a system, video measuring systems are available at different levels of performance and price, and systems with similar measurement ranges or accuracy specifications may not provide equal performance.

Because these systems are used to monitor and maintain a manufacturers' quality, it is important to be able to trust their measurements, so price alone should not be the determining factor on what to buy.

An inexpensive system may not perform as well over time.

It may drift out of alignment or be made with marginal components, leading to unplanned service and repair costs over its lifetime.

The best way to decide on a system is to see how well it measures your part.

When considering the specific system to buy, think about how important those measurements are to the cost-effective manufacture of your products.