Accuracy is defined as the quality of being close to the true value of an object. Knowing the accuracy of your 3D scanner is important because the information derived from 3D scans are used in decision making, whether it is used for quality inspection, design, or reverse engineering. As discussed in the previous blog post, The Basics of 3D Scanning and Tips to Get Optimal Results, checking the accuracy of the 3D scanner after calibration helps to determine whether you are getting the optimal level of accuracy the 3D scanner is capable of achieving. Usually 3D scanner manufacturers list the accuracy specifications of their products in their brochures or on their website.
To determine the accuracy level of a calibrated 3D scanner, you need to scan an object of known measurements. Some of these objects include:
- Known geometric shapes made out of stable materials. These include items such as tooling balls or gauge blocks. For those who need the utmost accuracy, use a certified artifact. Our team uses several different NIST certified geometric artifacts and certified gauge blocks. (PHOTO LEFT)
- Digital Calipers that have a locking feature. (PHOTO RIGHT)
- An accurate ruler, preferably ceramic or metal
Conducting an accuracy test requires you to take several scans of the object for a specific field of view. It is common to have different accuracy levels for a particular field of view so it is important to take multiple scans from different positions within the scanner's field of view when conducting the test. The more 3D scans you use, the better the test results will be. Once you have a number of 3D scans, compare them against the true measurements of the object using 3D scan data post-processing software (ie. Geomagic, Rapidform, Leios). This process is known as deviation analysis.
This case study quickly outlines the basic steps of deviation analysis. In this example, tooling balls are used to check for accuracy.
The tooling balls come with a certificate that specifies its measurements. Each sphere has a known diameter size and a known distance between the two spheres. Once several 3D scans were taken, we imported all the scans into the software.
In the software, we selected the first raw scan and apply the Sphere Feature. This function determined the estimated diameter of each sphere and the center distance between the tooling balls (x, y, z coordinates) in the 3D scan.
Once the software estimated the diameter and the center coordinates, we compared these numbers against the known measurements. To get a larger sample size, repeat this step with a different 3D scan. Once we’ve done this several times, we can calculate the average deviation of the 3D scans we’ve made.
If you find you are not getting the same accuracy level as specified by the manufacturer, you would need to re-calibrate the 3D scanner and check the accuracy level again. This ensures you are getting the optimal level of accuracy that is expected of your 3D scanner.
This blog post was previously posted on Tue, Mar 1, 2011 on 3D3 Solutions blog. LMI Technologies acquired 3D3 Solutions on May 1, 2013.
Posted by Thomas Tong