Digital technologies/3D Scanning
Introduction
3D scanning is the technology of creating a 3D mesh from some mix of scanning techniques. Most 3D scanners use structured light or laser triangulation to scan what they see and use photogrammetry to create the 3D mesh by combining all the 2D scans together. The mesh can be used to create 3D models of various things and applications. For example, scanning a limb to create a custom prosthetic or scanning components to design a custom case, etc. At the makerspace we have the EinScan Pro X2020 which uses laser triangulation to take 2D scans of an object, and its paired with the EinScan software which complies everything together. Both require the other to work properly but the raw scan files can be imported to other post-processing software as needed.
Applications of 3D scanning
3D scanning is a revolutionary technology which can be used on a day-to-day basis where you can easily scan a part, convert it to a CAD model and make necessary adjustments if required. Does a certain part not fit perfectly to an assembly? Just do a quick scan of that part, convert it to a CAD model, and resize it to perfection.
3D scanning has become increasingly popular in reverse engineering, which can be employed to make a 3D model of an existing part. Reverse engineering is beneficial in many ways since it helps to examine older creations and inspire novel ideas based on them, by designing the CAD model.
Elsewhere, 3D scanning is an efficient tool when prototyping, where you can quickly obtain the dimensions of a 3D object and, hence, create other prototypes. 3D scanning is thus a great tool to work with for research purposes.
Nowadays, 3D scanning is a technology present in numerous industries such as in the space industry, to scan space rocks, in the construction industry for quality control and entertainment industry, mostly used in virtual cinematography.
How does it work?
The EinScan Pro x2020 can be used in a few ways. There's the fixed scan mode, where the scanner remains stationary, and the object is rotated in front of it to get all the sides.
Or it can be used in handheld mode, with the scanner being moved around the object and the object remaining stationary.
Either way, the software will take all those pictures and create point clouds which can be used to create a 3D mesh of the object. Within the scanning software, light post processing can be done to modify the mesh as needed before exporting it as whatever file type is needed for the next step in the process. This step varies depending on what you are trying to do with the 3D scan.
Handheld scanning is more useful when the object to be scanned is too big to fit the turntable or has complicated features. The following video indicates in what cases the hand-held scanner will be more practical.
Files Types
3D scanning by default creates point clouds when scanning. Point clouds are a collection of data points in space, together forming an object. These point clouds are then connected to create a mesh. The raw point clouds can be brought into other software's for post processing, but normally the resulting mesh is the file that is imported and exported between software's. However, directly from the EinScan software, as well as most 3rd party software's, the mesh can be exported as other file types like .asc, .stl, .ply, .obj, .3mf and more. For example, to directly print the model it should be exported as an .stl. However if more post processing was to be done or modified in another CAD(computer-aided design) software it could be exported as a .3mf file.
3D Scanning in MakerSpace
Our 3D scanner
The EinScan Pro 2X 2020 handheld scanner has a scan area of 150 × 120 mm—250 × 200 mm and works at a minimum point distance of 0.2 mm. It can catch accurate details of either small and medium-sized objects: from 3 to 100 cm in tripod mode and from 30 to 100 cm when handheld.
Specifications of the scanner
The following video demonstrates how to use GeoMagic Wrap 2021 to edit your 3D scanned object. Make sure to save your file as STL or 3MF to be able to open it in the software.
| Spec | Handheld HD Scan Mode | Handheld Rapid Scan Mode | Fixed Scan Mode with Turntable | Fixed Scan Mode w/o Turntable |
|---|---|---|---|---|
| Single Scan accuracy | up to 0.045 mm | up to 0.1 mm | 0.4 mm | 0.4 mm |
| Scan speed | up to 10 fps;
up to 3,000,000 points per second |
up to 30 fps;
up to 1,500,000 points per second |
<1s | <1s |
| Point distance | 0.2-2mm | 0.2-2mm | 0.16mm | 0.16mm |
| Align modes | Markers, feature, hybrid alignment | Markers, feature, hybrid alignment
(for complex geometric objects) |
Turntable coded targets
alignment, markers, feature, manual alignment |
Markers, feature, hybrid alignment |
| Full-color scan | No | Yes
(Requires a Color Pack add-on) |
Generally, in the Makerspace, doing a 3D scan consists of choosing a scan type, preparing the scan in the software, prepping the object itself, and scanning the object. The resulting scan can be cleaned up at a basic level in the EinScan software and then exported to a stronger post-processing software for further processing.
Choosing Scan type
When the EinScan software is opened, a scanning method must be chosen. Either fixed or handheld.
A Fixed scan keeps the scanner stationary while the object is rotated on the supplied rotating plate. A Handheld scan keeps the object stationary, and it is the scanner that gets moved around to capture multiple scans from different angles. For smaller objects the Fixed mode should be used whereas for larger objects, the handheld mode should be used.
The following video(in French) demonstrates an example of using rapid scan.
Furthermore, there's handheld HD scan and handheld Rapid scan. The rapid scan captures less points per second but captures more scans per second so it will create a lower quality but more complete scan faster than the HD scan.
Overall, they are used in the same way. Stationary object, moving scanner.
Selecting Texture
After choosing a scan type, you will be prompted to select the scan's texture. The following picture demonstrates the dialog box that appears if you have selected Fixed scan. A non-texture scan will focus mainly on the object shape and will be grayscale or have a neutral color, while a texture scan will also bear the colors and other details of the scanned object as well as the shape.
The open global markers file will be addressed more in depth in Align mode->Global markers section.
If handheld scan was chosen, the following dialog box will appear. For Texture, it is similar to fixed scan, that is if you want to focus solely on the shape of the object select non-texture scan and select texture scan for objects that you would like to capture the details. While for mode of alignment, hybrid is using both markers and features present on the object to detect and scan the object.
Turntable settings
During a fixed scan, the object is placed on a turntable that automatically rotates so that the scanner can capture multiples scans from different angles. The number of steps in the full rotation is user set when prepping the scan. A smaller number of steps takes less time but may result in rougher scan quality (as it has less individual scans to take so less data to use). A higher number of steps will take longer but increase the quality of the scan (as there are more scans to help piece together the overall 3D mesh). A good rule of thumb is to start low around 6-10 and increase the steps as needed if the quality of the scan is too low.
When setting up a object on the turntable, make sure the object is centered on the turn table and is fully viewed by the preview of the scanner. Also verify that is it properly supported to the turntable and that it won’t slip as it turns. If it will, use some sticky tack to temporarily stick the object down.
Also, make sure that the object is placed in the scanner field by checking what is shown on the software camera when you are moving the object. Ensure that the + sign is aligned with the center if the turntable and place the object in the center as well. The following picture demonstrates how the scanner is aligned to the center of the turntable, ensure that the + sign is inside or close to the red box.
Align mode
Turntable coded targets
Recommended for objects that fit the turntable and that you would prefer not to have targets on the actual object. This option makes the scanner detect common markers on turntable to calculate the object’s new location as the turntable rotates.
This alignment mode makes it unnecessary to have markers on the object and also if it happens that the distance between the camera and the turntable fluctuates, you are not required to recalibrate the scanner.
Features
For an object that is too big for the turntable, and you would not be able to use markers, the features alignment method should be chosen. In this case, the software recognizes features on three consecutive early scans and then calculates the distance between the scanner and the turntable.
When choosing this option, ensure that the object is in a stable position on the turntable and the primary view has enough features to start the scanning process. Plain objects without striking features would not be recommended to be scanned with this alignment mode.
Markers
Markers are used when the camera cannot detect the targets found on the turntable. You should use stickers or marker dots on the object. See Prepare the Object->Markers for more information.
Global markers
This option is quite similar to using markers as an alignment method. When creating a new project, import a file with the global markers, known as global registration file or a .p3 file, which can be saved from a previous project using fixed scan mode or hand-held mode.
When choosing the markers or global markers align mode, you will be required to stick an adequate number of markers to ensure that the scanner can match consecutive scans, if some markers have already been scanned and imported. It may be required to increase the turntable steps to ensure that the scanner follows these consecutive scans, and also to be sure that the scanner detects the markers.
HDR (High Dynamic Range) Settings
The HDR (high dynamic range) settings alter the level of contrast that the scanner will use to detect an object as it flashes light during the scans. A lower HDR won’t pick up as much contrasting details. There is a preview in the program that shows what will most likely get picked up shown in red that can be used to ensure the object being scanned is properly captured. In short, if there are more colors, a higher HDR should be chosen to ensure they all get properly scanned.
Prepare the Object
Reflectiveness
Some materials are too reflective or too dark to properly be scanned. They either bounce off too much of the light, even at low HDR settings, or still absorb too much light, even at high HDR settings. In either case, to combat this, the object can be dusted off with some talc powder. The talc powder helps balance out the reflective properties of the material, from extreme to extreme. To prepare a reflective object to scan, you should first use a brush to cover its surface with talc powder. Use the talc and brush found in the cabinet at the 3D scanning setup and apply a layer of talc powder on the object, make sure that the object shows up on the software once the talc powder has been applied. The following picture demonstrates how a reflective object appears in a scanner field of view. Since, it is not shaded in red, it means that the scanner could not detect the object and hence the turntable appears empty.
The following animation depicts how to reduce an object reflectiveness by adding talc.
Now, as seen in the picture below, the key is shaded red, hence indicating that the scanner can detect the object.
Markers
If you are planning to take multiple scans (explained below) or are scanning on an object with large flat surfaces, simple geometry, or otherwise not enough features on it, it can be helpful to add temporary markers on the object to aid in aligning the scans. Some examples would be small stickers or marker dots/markings as shown in the following picture:
The following example shows an example where markers are useful. The object shown is dark, has some parts which are reflective and a variety of features. The markers helped to detect the chair back and seat, but for the reflective part in the middle, the markers were not enough, and hence talc was applied to help the scanner detect this part.
The result obtained is shown below:
The following video gives a more in-depth explanation of using markers before scanning an object.
Taking a scan
Once the settings have been set up and the object is ready to be scanned, the scan can be started by pressing the start scan in the top right corner.
During a fixed scan, do not move the scanner or object. Once the rotations are complete, you can clean up the scan and move on. During a handheld scan, do not move the object and use the onscreen display to make sure you are within the correct scanning distance.
Immediately Cleaning up a scan
Once the scan is complete, there is an initial option to roughly clean up the scan. This is a quick chance to manually get rid of the obvious “noise” picked up by the scan like the turn table, other objects in your scan that are not the target object etc. To get rid of these portions of a scan use the tools available to circle the noise and delete it. The scan can be rotated to get to every angle.
To remove unwanted features of a scanned object, hold the Shift key and use the left button of the mouse to select unwanted pieces and click on Delete.
Once a part is selected, it should appear in red as shown below:
Watertight and Unwatertight model
After a scan, you can generate data points and then select mesh model to generate a 3D model. You will be prompted to select Watertight or Unwatertight model. A Watertight object is an object that is completely sealed, and no gaps are present, select this option if you have scanned an object in all directions.If Watertight is chosen for an object that was not scanned enough in all directions, the software would try to create the missing parts that were not scanned to complete the seal and hence compromising with the object actual structure.
The following demonstrates what happen if you select watertight model and not scanning enough to generate the complete object.
The chair was only scanned on the top, hence Unwatertight was chosen to make sure that the scanned model represents the actual object to a tee.
After creating a mesh model, the model below is obtained. More post-processing tools such as removing holes and sharpening are found on the right.
Taking multiple scans
Sometimes, 1 scan is not enough. Multiple scans might be needed to fully capture the details of an object. For example, if the base of an object has geometry that is hidden when it is lying flat. In this case, a second scan, with the object in a different orientation, would be needed to capture those hidden geometries. To take another scan, create another project within the same group and then do the same cleaning process as the first scan before aligning the 2 scans into 1.
Aligning multiple scans
To align multiple scans, the align feature can do it automatically, or the user can set 3 points on both scans that are the same to align the scans. Unless the object being scanned is very distinct, manual points should usually be used to align the scans. Once aligned, the mesh should be more complete.
As previously mentioned in preparing the object, temporary markers like small dot stickers or marker dots can also be added to aid in finding manual points to align the scans.
The following video depicts a project requiring two scans and aligning them afterwards.
Exporting the scan
Once scanning is done, the mesh can be exported in a few different file types, such as.asc,.stl,. ply,.obj,.3mf. These can then be brought into other post-processing software to manipulate and dimension further, or even directly into fabrication software to 3D print, for example.
To export a scan, select the save icon found on the right as shown below.
Post-Processing in other software
In Makerspace, there is the software GeoMagic that lets the user import a mesh. In GeoMagic, the user can manipulate the mesh with more control and dimension to make sure it is the correct size. GeoMagic can also help in ensuring a complete mesh by filling in any missing areas of the scan or replacing shapes of the mesh. Its only drawback is that it is not free, so if you are not in the Makerspace, you cannot use GeoMagic. However, there are various free, open-source mesh editors that do similar functions. For example, Meshlabs. MeshLabs is open source and free to download. It has similar capabilities to GeoMagic including dimensioning known shapes, cleaning up the mesh and optimizing the numbers of polygons, as well as making it a watertight mesh. The following video demonstrates how to use GeoMagic Wrap 2021 to edit your 3D scanned object. Make sure to save your file as STL or 3MF to be able to open it in the software.
The following video demonstrates how to use GeoMagic Wrap 2021 to edit your 3D scanned object. Make sure to save your file as STL or 3MF to be able to open it in the software.
[5]