FAQ
In this area, we answer the most frequently asked questions related to “measurement using vision measuring systems”.
Just click on one of the questions in the list to read the corresponding answer.
If a question you are interested in is not listed, please send an e-mail to the vision measuring experts at Mitutoyo’s. Your question will be answered individually and, eventually, taken into the FAQ list.
Why does Mitutoyo not employ high-resolution color cameras in its Quick Vision instrument?
What is the difference between a zoom lens and an automatic magnification changer?
Which are the advantages and disadvantages of a telecentric optical system?
How accurate is a measurement on the image – without instrument movement?
Why does Mitutoyo not employ high-resolution color cameras in its Quick Vision instruments?
New digital camera types entering the market are tested by Mitutoyo’s research and development department KDC in Japan in regular intervals. Up to now, none of these high-resolution cameras proved to enhance the measuring results, but instead even slowed down the evaluation process.
This originates in the cameras’ performance at edges, i. e. in those areas, where color or brightness of the image suddenly change. Edges are detected by the software as several points along the measuring path (perpendicular to the edge). For example, in pitches of 10 pixels, one point on the edge is determined.
From these measured points an ideal straight line is averaged and output as result. Additionally, the span, in which the measured points scatter around the averaged straight line, is output as an indicator of the quality or straightness of the edge.
Low-resolution cameras represent the edge slightly coarse (or better: stepped). Cameras with higher resolution produce smaller pixels and thus also smaller steps along the edge. Theoretically, the image of the edge should be clearer and better defined. In reality, however, the smaller pixels generate a flickering on the edge, i. e. some pixels continuously change in brightness – since they are very small and lying exactly on the edge line. The brightness of the coarser pixels produced by cameras with lower resolution instead remains stable.
The flickerung on the edges makes the detected measured points vary at least as much as points detected in a coarser – but stable – grid, or, in most cases, even more. Consequently, the average result is not better, but mostly even inferior.
To balance the loss caused by the relatively coarse grid, Mitutoyo employs a microscope optical system in front of the CCD chip which captures the image. This microscope optical system enlarges the edge line to an extent that one pixel has a size of only a few micrometers - thus minimizing the sum of the positional errors caused by the pixel in relation to an existing form deviation of the edge to a negligible degree.
In addition, the coarse grid is always refined arithmetically before evaluation. By this sub-pixeling smooth transitions in the video image displayed on the monitor are achieved – and the disturbing pixel grid will be avoided.
What is the difference between a zoom lens and an automatic magnification changer?
In a zoom lens, multiple lenses in the objective are shifted in correlation to each other leaving the resulting distance between object plane (on the workpiece) and image plane (on the CCD chip) unchanged, whereas the magnification factor of the complete system is changed correspondingly.
The automatic magnification changer, however, is based on microscope optics. This means, here, the image of the object plane is firstly enlarged by a fixed lens and secondly converted to a parallel bundle of light. This bundle of light is converted into an image (on the CCD chip) in an arbitrary position by using another lens – the tube lens – and at the same time enlarged again. In the automatic magnification changer employed by Mitutoyo three tube lenses featuring different magnifications are housed in a motorized revolver. The user can position each tube objective into the optical path as required to obtain different total magnifications.
The advantage of this method is the fact that high-quality microscope lenses can be employed which eliminate all possible lens aberrations. This can only be achieved by an elaborate combination of different lenses in diverse forms and made from varying sorts of glass. Therefore, a zoom objective which must be equipped with movable lenses can not provide a correction of equivalent quality. In other words: The image generated using an automatic magnification changer is clearer and sharper than the image generated by a zoom objective. Furthermore, most fixed magnification lenses feature higher optical resolution, thus producing clear images even of small details that could not be detected by a zoom lens at all.
Which are the advantages and disadvantages of a telecentric optical system?
For measuring technique, the telecentric lens provides a substantial advantage. Due to the configuration of these lenses no perspective distortion is generated. When normal lenses are used, the image size corresponds to the real workpiece size only directly in the focus plane. Is the object positioned in front of this plane, it will be imaged too large, and if it is positioned behind the plane, it will be imaged too small. If the object is measured in the image, the dimension detected and output will be wrong. In comparison, telecentric lenses always represent the object true-to-scale, no matter from which distance – measuring errors can not arise.
This large advantage is, on the other hand, accompanied by two serious disadvantages. Firstly, due to their principle telecentric lenses always have a smaller numerical aperture. This is the parameter which determines the optical resolution of lenses. The higher this value, the smaller details can be represented by the optical system. This means that telecentric lenses always have a relatively poor optical resolution - especially in comparison to other microscope lenses.
Secondly, telecentric lenses feature a large field of focal depth. Within this area, the image will constantly have the same sharpness. If this area is relatively large – as it is with telecentric lenses – the height of objects can not be stated exactly by focusing the image. In other words, optical measurements in vertical direction are not possible or very inaccurate. Thus, a three-dimensional measuring instrument is turned into a two-dimensional one.
On the other hand, microscope lenses feature a very small depth of field, so that actually any element not lying exactly on the object plane and therefore represented perspectively distorted, is at the same time represented unsharply. Erroneous (perspectively distorted) measurement can therefore be avoided by simple filtering of unsharp image areas.
For this reason Mitutoyo always employs conventional microscope lenses in three-dimensional vision measuring systems.
How accurate is a measurement on the image – without instrument movement?
This question is very difficult to answer. Basically, the possible error depends on the pixel size, of course. This size, in turn, depends on the magnifying optics installed in front of the CCD chip. The higher the magnification factor, the smaller the represented image point. The video image is always digitally processed – by sub-pixeling – to determine the brightness distribution within an image point, thus improving the actual resolution of the image arithmetically. This means that an exact statement on the quality of a measurement performed on the image alone can not be given.
It is possible, however, to measure a master workpiece on the image and to calibrate the image point size according to this specimen. Subsequent measurements will then be referenced to the dimensions of the master part - each part will be compared to the master part size. Such proceedings generally achieve higher accuracy levels – however, these can not be stated as numerical values.