QMT Features: June 2009
A multisensor vision
Ultrafast microprocessors and advanced CAD-software have combined to bring vision and multi-sensor measurement into the mainstream of manufacturing. By Marc Stalker, Wilcox Associates, Hexagon Metrology.


1. WhitePoint Microprobe: A multisensor probing options supported by PC-DMIS Vision software.(Photo courtesy of Mycrona)

Vision and multisensor measurement is one of the most exciting and fastest growing segments of the metrology hardware and software marketplace. The reason for this is quite simple. Ultrafast microprocessors and advanced CAD-software have combined to bring vision and multi-sensor measurement capabilities and their advantages into the mainstream of manufacturing.

Not long ago, technical specialists in research labs were the only users of vision and multisensor measurement equipment. Today these systems are operated in manufacturing environments and they are programmed using software very much like the software used to program and operate CMMs. People who already know how to use CAD-based CMM programming can learn how to program a vision or multi-sensor system with only a little additional practice.

While the programming and operating of these systems is straightforward, many of the adjustments and calculations that occur within the software during the programming and measurement are very complex and sophisticated. These take place automatically behind the scenes so that the user can focus on what he is most interested in, dimensional measurement results and their analyses.

 In the past, if you didn’t know a lot about focus, lighting and magnification, you could rule out vision measurement as a practical, everyday, option. Recently some vision metrology software developers have brought vision within easy reach the common user by incorporating this esoteric knowledge in the software itself, so that these parameters can be adjusted automatically with little operator intervention.

This new breed of vision and multisensor software allows users calibrate the illumination of vision machines so that they provide extremely linear and accurate measurements over a wide range of illumination intensities and material reflectivity.   Additional functions automatically adjust lighting intensity on the fly to improve contrast.  Special algorithms evaluate contrast in the area of interest and then make adjustments if needed.  Some measurement software can even adjust lux values to compensate for the gradual loss of illumination intensity over time.
Controlling focus is another area of improvement. 

Again, special algorithms evaluate edge strength and decide if it in necessary to refocus to improve accuracy. This functionality improves measurement cycles by selectively eliminating the need to refocus every time the camera pauses over a new area of interest. In addition some software allows for calibration of focus on individual machines for optimal measurement speed and accuracy. This can be particularly useful when the same parts may be measured on more than one type or brand of vision equipment.

Advanced Edge Detection
Another formerly, but no longer, esoteric problem for those wanting to measure with vision was identify edges on the features of interest. Users had to understand how to manually invoke a variety of filters to find edges that might be ambiguous to the camera even though they are clearly visible to the naked eye.

Again some software alternatives have come to the rescue by providing  user-selectable edge selection algorithms.  Wilcox Associates’ PC-DMIS Vision measurement software has a number of these. The user specifies the type of edge he is looking for and the software applies the algorithm that is most likely produce the greatest edge strength (a function of the contrast between adjacent gray scale pixels in the image) for the evaluation in question.

With this sort of intelligence built into the software, the user does not have to become an expert in manually applying various filters for selecting edges. The software does the heavy lifting (computationally speaking) and he can get on with the job of measuring his parts.

Measurement throughput
Measurement throughput is a great concern in the world of production manufacturing. Vision systems are great and gobbling up and evaluating great swaths of territory within the field of vision. However, when they are used to capture discrete points from 3D surface, progress may be much slower than with a tactile probe because the camera must stop, stabilize and refocus.

New software and hardware enhancements are being developed to greatly improve the measuring efficiency of vision and other non-contact probing alternatives. However, something you can take advantage immediately is vision and multi-sensor software that takes advantage of dual and quad core CPUs so that imaging and image processing can take place simultaneously to significantly speed up measurement results.

Programming throughput is also of great concern to users who want to employ vision and multisensor capabilities in short run and product development environments where programs must be continually written for parts and their subsequent variations. Parametric programming techniques, long available to CMM software users, allows for the rapid programming of families of similar parts simply by changing values in parametric tables.

This type of programming is particularly important for products where there are many variations to meet specific individual requirements. For example, parametric programming of multisensor measurement equipment is saving some medical parts manufacturers man-weeks of programming time for products like prosthetic devices. 

White light sensors recently introduced for use with multisensor measurement equipment, can capture data six times faster than lasers and with greater accuracy. Unlike lasers, white light probes rely only on the evidence of the signal (rather than its power and amplification). So there are no significant issues relating to angle of incidence or surface reflectivity and white light sensors are not thrown off by such things as colour and texture variations; translucent, transparent or highly reflective surfaces; or variable ambient lighting conditions.

On multisensor machines, users are currently limited to using white light spot probes for capturing point data from surfaces.  Some multisensor software allows white light probes to operate in a manner identical to tactile probes for linear, patch and rotary scans. So these probes are exceptionally effective tools for capturing high-resolution data from free-form sculptured surfaces. The rate of data capture is limited only by the computer’s ability to process and store the collected data.

Down hole microprobing
White light probe technology is evolving rapidly. Recent developments include the introduction of a tiny a WhitePoint Microprobe (WPM) capable of measuring inside very small holes.

The WPM has an interchangeable probe tip that connects to a flexible optical fibre.  The probe tip and optical fibre are connected to an independent sensor interface that acts as the white light source and processor for the data being collected. The optical exit of the extremely small probe tip (35 µm) is aligned exactly 90° (X-Y-plane) to the 360° rotating vertical axis.

With a resolution of 1 nm and the high-accuracy positioning of an ultra-precision multisensor system, the WPM allows for fast measurements on very small internal and external geometries with a measurement accuracy of up to 200 nm.

This article may have led you to the conclusion that you have to purchase a multisensor measurement system to do tactile and non-contact probing on the same machine. That is not necessarily the case.
Another way to check features that don’t always lend themselves to measurement with a tactile probe is to equip a conventional CMM with an articulated wrist and a vision probe. As with a touch trigger probe, the measurement software allows the programmer to orient the probe normal to the feature it is measuring. Since many shops already have the CMM, they only need to buy the vision probe.
CMM-based vision probes are currently limited as to the amount of magnification they can provide. Therefore, they are not yet appropriate for very small features requiring ultra-high resolution measurement. That is likely to change however as CMM-based vision probes evolve.

The takeaway from all this is that vision and multisensor measurement is no longer too specialized or difficult for the average user to consider. So if you have any of these problems:

  • Some features may be damaged or distorted by contact,
  • Features to small too be entered or accurately measured with standard probe tips,
  • Features not accessible to hard probing,
  • Need to acquire many simultaneous data points;

then you and vision or multi-sensor measurement may finally be right for each other.

email: Marc.Stalker@hexagonmetrology.com www.hexagonmetrology.com

  
You can now view all QMT Magazine issues on your favourite tablet or smart phone.
Download the free Quality Manufacturing Today App from the Apple iTunes App Store or from QMT Magazine on Google Play.

Rob Tremain Photographer
www.4exposure.co.uk
slideShow
Click above to see full page display and links to QMT articles.
Untitled
Nikon logo
Prodim logoi
Control logo
Mitutoyo logo
Vision 2016 ad
Creaform logo
TCT Inspex 2016 logo
Hexagon Logo 2