Introduction

SpatialAnalyzer (SA) and Geomagic Control are two powerful pieces of measurement software which can be used to scan freeform surfaces to capture point cloud data and mesh the data to create a polygonal model. Traditionally SpatialAnalyzer has been the software of choice for probing rather than scanning, with Geomagic packages being designed specifically for scanning / acquiring large amounts of data quickly.

In the latest release of SpatialAnalyzer (2016.02.18) improvements have been made regarding point cloud handling and new meshing capabilities had been added. The purpose of this case study is to compare the capabilities of the two software packages by scanning an automotive part and meshing the resultant point clouds.

Data Acquisition

The surface of the sheet steel component was scanned in both SpatialAnalyzer and Geomagic Control using a Romer 75 series scan arm.

Speed of acquisition: Both software packages were able to scan the component and complete the inspection procedure, however Geomagic control was the quicker regarding data acquisition. Geomagic almost instantly displayed the scan data in the graphical view whereas SA had a slight delay in showing the captured data on the screen. SA also occasionally ceased collecting data mid scan, meaning a second pass would be required.

Graphical representation of the data on screen: The graphical view of the point cloud data collected in SA was not as clearly displayed as in Geomagic. In SA it was very difficult to see if areas had been missed due to the point cloud appearing quite fine. In Geomagic this was not an issue as any gaps were shown clearly indicating a second pass was required to fill them in.

Point Cloud Data Alignment

When aligning the data two different methods were available.

Method 1: Aligning the top and bottom scan to one another without using the CAD data. In both software packages this method didn’t work as well as expected. In Geomagic, an N point alignment could be used to roughly align the two sets of scan data, this allows the user to pick matching points on overlapping data in order to roughly align the two scans. A best fit could then be performed using one set of scan data as the test and the other as the reference. This best fit took a long time to perform and didn’t seem to align the data perfectly. This could be due to the very thin wall thickness of this particular component. In SpatialAnalyzer there is no method of performing this type of alignment therefore aligning scan data with no nominal CAD is not an option in this software.

Method 2: Aligning the top scan and the bottom scan to the CAD individually – This method seemed to work much better in both software packages. In Geomagic this procedure is exactly the same as the one described above, however this time the reference object will be the CAD model. An N-points alignment can be used to get the data close. A best-fit can then be performed. This method is very quick and effective, giving accurately aligned data in next to no processing time. In SA the process is slightly different. There are 2 options given when best fitting point cloud data to a nominal CAD model, neither of which are very quick or effective.

To roughly align the data ‘quick align clouds to CAD’ can be used. This option allows you to select inspection points on the CAD and corresponding points on the point cloud. The main problem with this operation is the graphical view of the point cloud is not very clear. It is not easy to see whether you are selecting the correct area on the cloud. Once the data is roughly aligned a ‘point clouds to object best fit (or relationship)’ can be performed. This process is effective yet time consuming. In this particular example the best fit took over 20 minutes. When using SpatialAnalyzer ideally the instrument should be in the correct alignment before taking the scan.

Meshing / Creating a Polygonal Model from the Point Cloud

Once the point clouds were aligned a polygonal mesh was created for each in the two different software packages for comparison. This was an easy and quick operation in both programmes using just a few clicks. Below we compare the quality of mesh in each software:

Overall the quality of both meshes was very impressive. The mesh created in Geomagic Wrap seems to be slightly more complete with less holes and defects, however this could be because of a more complete scan due to a better graphical display at the scanning stage.

Both polygonal models have some defects. Above we see imperfections in the mesh on the same lip of both polygonal models. The Geomagic mesh shows imperfections that look like a ripple in the surface whereas we see on the SpatialAnalyzer mesh these imperfections show up as holes in the model. Geomagic Wrap has tools for working on these errors, whereas SA currently does not.

Conclusion

This case study has found that both SpatialAnalyzer and Geomagic Control are an effective and powerful software packages when it comes to scanning and meshing to create polygonal models for reverse engineering. Geomagic Control had the best results when it came to speed of data acquisition and quality of graphics whilst taking the scan. Geomagic also came through on top when it came to cleaning up the point cloud and deleting rogue data.

Alignment of scan data without a CAD model was an option in the Geomagic software however it didn’t work as well as it could have. When aligning the data to the CAD model Geomagic was again the fastest software to perform the best fit, however both software packages worked effectively and accurately to give good results. SpatialAnalyzer seemed to take a little longer when performing operations such as aligning and best fitting of data.

The final output of both software packages was very impressive. Both produced a nice clean polygonal model which could be exported into a reverse engineering software for further work. The polygonal model produced by Geomagic appeared to be of slightly higher quality.

With SpatialAnalyzer traditionally being a software used for probing only, this case study has proved that its scanning and meshing capabilities have improved dramatically in this new version. Geomagic Control is still the best scanning software of the two, but SpatialAnalyzer is making steps to catch up.