Vision, Laser and Combo scanning compared
ASCAND Vision, Laser and Combo scans are different ways of obtaining and interpreting geometric evidence. They are not three quality levels.
Vision constrains object volume from camera observations, including silhouettes. Laser triangulation derives surface samples where a projected line reaches the object and the camera detects it under the modeled geometry. Combo obtains related evidence for both paths, reconstructs it independently and evaluates how the resulting geometry can contribute to a combined representation.
The right method is the least complex supported method that can observe the geometry required for the intended result.
Start with the object and the purpose:
- Which surfaces must the result represent?
- Does the outer silhouette contain enough information?
- Are important recesses visible to both the laser and camera?
- Will the surface provide usable optical evidence?
- Would a second evidence path address a specific weakness?
- Is an unseen underside really a method problem, or does it require another object orientation?
These questions separate a defensible choice from the assumption that more processing produces a better model.
The methods differ in what they observe and reconstruct
All three methods share ASCAND’s controlled acquisition foundation: a steady camera observes an object rotating on the coded turntable. What differs is the evidence used to infer geometry and the responsibilities that follow.
| Comparison | Vision | Laser | Combo |
|---|---|---|---|
| Primary evidence | Object boundaries, silhouettes and applicable image evidence across the rotation | Detected positions of the projected laser line | Related Vision and Laser observations |
| Geometric principle | Multiple views constrain occupied volume; silhouette carving can form a visual hull | Camera rays intersect the modeled laser plane to form surface samples | Independent geometry is compared and reconciled according to its support |
| Additional hardware | Standard camera-based capture path | Compatible, aligned Laser Extension Kit | Compatible Laser Extension Kit plus supported Vision and Laser evidence |
| Characteristic value | Coherent outer-form evidence without requiring a projected laser line | Local surface evidence that can describe visible depth changes beyond an outline | Complementary sources can support different regions or reveal disagreement |
| Characteristic boundary | Hidden concavities that never affect an outline remain unavailable to silhouettes | Laser shadow, camera occlusion, alignment and surface response limit sampling | Shared blind spots remain; conflicting or weak evidence still needs judgment |
| Intermediate geometry | Vision-derived volume, points or surface representation | Laser-derived point cloud | Separate Vision- and Laser-derived geometry before combination |
| Evaluation still required | Segmentation, outline fidelity, hidden shape and intended-use checks | Line detection, outliers, coverage, alignment and scale checks | Source alignment, agreement, conflict, gaps and contribution checks |
These are method responsibilities, not guaranteed results. Geometry, capture quality, calibration assumptions and lighting influence what remains usable.
Multi-Scan is not a fourth sensor method in this comparison. It captures the same unchanged object in additional physical orientations and later registers overlapping reconstructions. A Vision, Laser or Combo result can still need Multi-Scan when the turntable contact area or another surface remains hidden.
Understand Multi-Scan separately
Choose Vision when observable outer form is the main evidence
Vision is a strong starting point when visible boundaries describe the geometry that matters.
In silhouette-supported voxel carving, each segmented view separates object from background. The observed outline defines a visual cone: the set of 3D positions that could have produced that silhouette. Processing begins with a bounded volume and removes voxels that contradict one or more views. The survivors form the visual hull—the largest volume consistent with the available silhouettes.
Vision can establish a coherent outer volume without relying on a laser response from every surface. It can be appropriate when:
- the outer form is the primary geometric requirement;
- the object remains stable and clearly separated from its background;
- important features change the outline across the rotation;
- camera-only acquisition is required;
- a continuous starting representation is more useful than sparse local samples.
Its limitation follows from the same principle. A concavity that never changes an observed silhouette does not contradict the full visual volume, so silhouette evidence cannot carve it away. Deep recesses, internal cavities, undersides and self-occluded surfaces may remain absent or appear filled.
Segmentation also matters. If a boundary includes background, useful volume may remain. If it removes part of the object, valid volume may be carved away. A naturally closed Vision result is therefore not proof that the shape is correct.
Vision is not merely the “basic” method. When relevant geometry is well described by outer form, its evidence may fit directly. Laser or Combo adds value only if it addresses geometry Vision cannot represent sufficiently.
Learn how Vision and voxel carving work
Choose Laser when visible surface samples must go beyond the silhouette
Laser scanning is useful when important visible depth changes cannot be inferred from the outer boundary alone.
The Laser Extension projects a line whose light forms a plane in space. Where that plane intersects a visible surface, the camera observes a laser trace. Each detected image position defines a camera ray. Intersecting that ray with the modeled laser plane yields a candidate 3D surface sample. Repeating the observation around the rotation builds a laser-derived point cloud.
This can provide evidence for visible grooves, recesses, relief or changing depth that does not alter the silhouette. It may suit a project when:
- local surface shape matters more than a visual hull can express;
- the projected line can reach the relevant region;
- the camera can observe the surface response;
- the object’s material and surroundings permit reliable line detection;
- the compatible laser hardware can be aligned and operated under verified instructions.
Laser triangulation has a dual-visibility requirement. The laser must illuminate the surface, and the camera must see the response. A recess may be visible to the camera but shadowed from the laser, or illuminated by the laser but hidden from the camera. Neither case provides a usable triangulated sample.
Surface behavior also matters. Reflections can displace or duplicate the apparent line. Dark surfaces can weaken it. Translucent materials can scatter light below the visible surface. Ambient illumination can reduce contrast. These are evidence conditions, not absolute lists of scannable and unscannable materials.
Alignment and calibration assumptions connect image position to geometry. A clean-looking point cloud does not certify those assumptions or establish metrological accuracy. Operators must follow verified setup and safety guidance, then inspect coverage, outliers, scale and fitness for the intended use.
Understand laser triangulation
Choose Combo when the two evidence paths have a defined complementary role
Combo scanning is appropriate when Vision and Laser can contribute meaningfully different evidence about the same object.
Related laser-disabled and laser-enabled observations are acquired within the controlled camera–turntable relationship. Vision and Laser evidence then follows its relevant reconstruction path. Keeping those paths separate makes it possible to ask where they agree, where only one provides support and where they conflict.
Those states have different meanings:
- Agreement means both reconstructed sources support a similar local interpretation. It increases the available support but does not certify accuracy.
- Single-source support means one path contributes geometry where the other is absent or weaker. The contributing observation must still be evaluated on its own terms.
- Conflict means the sources imply incompatible geometry. This may reveal segmentation error, laser detection error, alignment problems, material effects or another failed assumption.
- Shared absence means neither method observed the region. Combination cannot reconstruct it from evidence that does not exist.
Combo is therefore evidence reconciliation, not a simple union of all points and not an automatic correction process. It adds capture, reconstruction, alignment, comparison and evaluation responsibilities.
Choose it when those responsibilities serve a defined need—for example, when broad Vision-derived outer form and laser-derived local surface samples provide complementary support. Do not choose it merely because a hybrid workflow sounds more advanced. If Vision already supplies the required geometry, Combo may add complexity without resolving a real limitation. If the laser cannot reach an important hidden surface, Combo does not change that visibility boundary.
Learn how Combo evidence is reconciled
Start with the limiting geometric feature
A method decision becomes clearer when it begins with the first feature that could prevent the result from serving its purpose.
1. Define the intended result
Identify whether the project needs recognizable outer form, visible surface relief, an underside, a digital reference, a printable starting mesh or another bounded outcome. “A complete scan” is not specific enough to guide evidence selection.
2. Mark the surfaces that matter
Separate required surfaces from those that are merely desirable. Note contact areas, cavities, undercuts and regions hidden behind other geometry.
3. Ask whether silhouettes constrain the required shape
If the important geometry changes the observed outline and the object can be segmented reliably, Vision may provide sufficient evidence. If a required recess never affects the outline, silhouette evidence alone cannot recover it.
4. Test the Laser visibility relationship
For geometry beyond the silhouette, ask whether the projected line can reach it and the camera can observe the response. If either path is blocked, Laser cannot sample that region from the current setup.
5. Define the reason for combining sources
Select Combo only when you can name the complementary contribution expected from each path. “Use everything” is not an evidence plan.
6. Separate method from orientation
If the missing area is the underside or turntable contact region, the limiting issue may be physical visibility rather than Vision versus Laser. A separate scan in another orientation—with reliable overlap for registration—may be required.
When the answer remains uncertain, use a controlled trial with a representative object and inspect the intermediate and final evidence. A test result is more informative than a categorical material or method promise.
Use the canonical object assessment
Different methods still share observation boundaries
Vision, Laser and Combo interpret different evidence, but none escapes the conditions under which optical observations are made.
| Shared dependency | What can go wrong | What to review |
|---|---|---|
| Stable camera, object and turntable relationship | Frames no longer describe the intended controlled sequence | Movement, vibration, object shift and framing consistency |
| Readable rotational reference | Observations may be associated with uncertain orientation | Visibility, obstruction, focus and continuity of the coded band |
| Suitable framing and focus | Required regions become cropped or indistinct | Full rotational envelope, sharpness and working distance |
| Surface visibility | Occluded geometry provides no usable observation | Camera view, laser reach where applicable and self-occlusion |
| Lighting and optical response | Boundaries or projected-line evidence becomes weak or misleading | Contrast, reflections, transparency, darkness and ambient light |
| Method-specific geometry assumptions | Reconstruction may be coherent but incorrectly placed or shaped | Calibration, alignment, scale and agreement with known evidence |
| Result evaluation | Successful computation is mistaken for a suitable model | Gaps, artifacts, false surfaces, completeness and intended use |
Adding another evidence source can reveal a problem or support a region, but it cannot guarantee that the underlying observations are valid. A successful processing status only means that computation completed. It does not establish that the model is complete, dimensionally validated, watertight, printable or suitable for a particular downstream decision.
Post-processing is also separate from acquisition. Point cleanup, registration, meshing, hole handling, surface repair, scaling and fabrication preparation may still be required after any scan method.
Review the shared capability boundaries
Select a method, then verify the evidence
Before recording, confirm:
- the intended result and required surfaces are defined;
- the selected method can observe the limiting feature;
- the object is stable and fits the supported capture relationship;
- the background, lighting and surface behavior support the evidence path;
- required optional hardware is compatible, aligned and used under current instructions;
- another orientation is planned separately when contact or occlusion hides essential geometry;
- the result will be inspected before it is accepted or prepared for downstream use.
Then follow the current tutorial for the selected workflow. Use troubleshooting for an observed failure, and compare the result with verified examples whose capture method and editing state are disclosed.
Method selection is the beginning of an evidence chain, not a guarantee at its end.
View verified scan examplesMulti-Scan captures the same physical object in more than one orientation. Multi Merge then registers and combines the independently reconstructed geometry.