The T3DMC team have previously been to demonstrations where many businesses were unaware of how photogrammetry works and in some cases what photogrammetry is at all.
However, following short demonstrations and discussions about photogrammetry by the team, businesses recognise how powerful it could be for their workflows going forward and for the flexibility of being able to measure very small and very large parts with greater accuracy.
What is photogrammetry?
Photogrammetry is the science and technology of obtaining reliable information about physical objects through the process of recording, measuring, and interpreting photographic images (2D imagery).
What tools are used?
There are different tools that exist within a photogrammetry system. A very basic principle of photogrammetry is ‘known information’ within a 2D image. Magnetic barcodes are applied to the surface to be scanned, which can then be captured by the handheld scanning device with integrated photogrammetry. A critical part of the process is using a scale. Typically, T3DMC would use calibrated ISO 17025 calibrated carbon fibre scale bars. Scale bars are a ‘known’ nominal distance that can be cross checked against and are the element that increases accuracy during a photogrammetry project.
Why is photogrammetry important?
It’s important to understand that photogrammetry captures a larger field of view than 3D scanning and because of this, it can more accurately capture reference markers used for 3D scanning services. It does this by increasing and varying the angle at which the camera sees the reference markers and the barcodes. As it’s a singular camera, which by definition is not locked to a stereo camera setup, it can move around more independently of the other camera position. This makes the calculation of all the points more stable.
Additionally, because the measurement volume is larger on a photogrammetry system, a larger scale can be introduced into the measurement – which are the reference carbon fibre scale bars.
Larger scale bars in a measurement of any kind will translate into a greater accuracy for that overall project because there is a larger known size within the measurement project. For example, if you use a one-meter part and have a one-meter scale bar, it will be a very accurate representation of scale. On the other hand, using a 10-millimetre scale bar in a one-meter project would not be very accurate. And this principle can be linearly extrapolated up to a much larger size.
There’s more to photogrammetry
If you want to simply 3D scan an object and add reference markers as you go along, then you can do this on a small component that fits within the natural measurement box of the scanner. However, if you add the reference markers on a bigger part, not only are you adding an unknown stack of error into the project, but you will also need to add a significant amount of dots to the project. This is to ensure you have a crossover of dots visible to the scanning device to move from one location to another.
Photogrammetry in practice…
If you wanted to scan the front and the rear of a car relative to one another, without photogrammetry you would need to apply dots the entire way along the surface, effectively joining the front to the back.
With photogrammetry you simply need to apply the reference markers in the region that you’re interested in scanning. In the example of a car, you can place some in the front and some in the back. In this instance you would be saving circa 75% of the dots needed to complete the project because you’re not having to stitch one area to the other. Photogrammetry, as a global measurement, takes care of that before you get to the scanning stage, further reducing your consumables usage. Photogrammetry also plays an important role in what we call measurement reference frames, also known as fixtures for holding parts.
If you have permanent points on a fixture, and it is in a stable environment itself, for example an inspection lab, then these reference points can be very quickly and accurately captured using photogrammetry which affords an accuracy of approximately 15 microns per linear meter. You can then set these as a reference project for the frame, meaning you’re in a position in which you can simply load parts on and off the fixture quickly and repeatedly, helping to increase process efficiency.
The takeaway?
Photogrammetry is an incredibly helpful and important part of 3D scanning and of the scanning services we provide in house – despite rarely being spoken about. It reduces to consumable usage and only adds to make project output more accurate. You can read more about how photogrammetry was critical in a project we undertook with Aston Martin in 2020.