Zivid last week launched its SDK 2.17, which comes with full Zivid 3 support and a new barcode reader. Zivid 3 XL250 is its sixth-generation 3D camera. The company said it extends the use cases for its cameras to applications that involve big robots working on large workspaces.
Oslo, Noway-based Zivid said it expects the camera and software development kit (SDK) to have an immediate impact on palletizing-related tasks in warehouse and logistics operations, and bin picking in manufacturing enterprises.
The company provides 3D cameras for industrial applications. Zivid spun out from SINTEF, one of Europe’s largest research organizations, in 2015.
Zivid 3 comes with two application presets
Zivid added presets to the 3D camera for specific tasks. For robotic depalletizing, it has two presets: one for the highest capture quality and one for the fastest capture speed.
The highest-quality preset for depalletizing is for scenes that may have strong ambient light and mixed SKUs of varied and challenging items. For example, these might be transparent bottles, shiny black polybags, or reflective soda cans.
Zivid said its fastest capture speed preset is still adequate for coverage on most items and can probably be used for most use cases. This setting has a less than 500 ms capture time. The company recommended using this preset first, and then switching to a higher-quality setting for challenging items.
3 specific presets designed for manufacturing
Manufacturing tasks have different considerations than depalletizing. Here, a robot may need to handle items made of metallic materials. This brings challenges such as interreflections among the parts themselves, with bin walls, and ambient light effects.
The three presets for manufacturing are:
- Specular: Zivid designed this preset for the shiniest metal parts, such as chrome tubes and reflective plates. It is intended to negate the reflection artifacts that are often encountered in such scenes, the company said. The capture time for this preset is typically 1,000 to 1,500 ms on a good computer.
- Small details: This preset excels at rendering small holes, raised areas, and intricate detail with high accuracy and precision, claimed Zivid. This can be especially important, for example, if CAD matching is used to identify parts, and such item features need to be clearly represented for the matching algorithm to work reliably. It also takes about 1,000 to 1,500 ms on a strong computer.
- Diffuse: This setting is the fastest preset for manufacturing, typically needing 600 to 900 ms. Zivid said it’s suitable for more matte surfaces that do not give off very specular reflections, for example, when picking steel billets from a large factory bin.
For users of slower machines, such as an integrated Intel GPU, the capture time can typically be two to three times slower, said Zivid. It has also made a few more presets available for other use cases, such as parcel handling with Zivid 3.
Zivid 3 can handle flickering lights
Zivid 3 has a new 2D imaging technique for applications where image stability and consistency are critical. This is available through the new setting called “Sampling Interval.”
2D images play a key role in robotic automation, as they’re commonly used to identify individual items in a scene. Zivid cameras provide their own illumination source using structured light. Other light sources are also usually present, so the company is working to suppress their effects as much as possible.
It has a range of techniques for doing so. With Sampling Interval, the system suppresses the effects of flicker and light-level fluctuation that can be seen with supply lighting that operates at 50 Hz or 60 Hz. This feature ensures consistent quality and color for 2D images, the company claimed.
Currently, this feature is only supported for Zivid 3 XL250, but the company expects to roll it out across its other cameras in the future.
Zivid can now detect 1D and 2D barcodes
Many robotic picking cells incorporate some form of barcode reading, and this is especially true in warehousing and logistics applications. With a new feature, all Zivid cameras support barcode detection and reading.
Users can find the barcode detector in the Zivid Studio by clicking ”show barcodes” under ”view.” The barcodes will be displayed in the 2D image view.
For best results, Zivid recommended using the barcode preset available under 2D presets. A range of common barcode types is currently supported, and users can specify which ones to look for.
The company noted that many things can affect the successful identification and reading of a barcode. This includes whether the code is occluded, damaged, at an angle, how far away it is, if it is in motion, and more.
Additionally, the effective spatial resolution of the camera matters, which again is a factor of sensor resolution, field of view, or depth of field.
Zivid said that out of its 3D cameras, the Zivid 2+ MR60 has the best barcode reading capability at the most pixels per millimeter. It can reliably read 10 mm barcodes, and it is possible to read as small as 7 and 8 mm barcodes. As cameras with larger field of view (FOV) are used, the reader works with progressively larger barcodes.
The company asserted that this new capability turns its camera into a “true multi-tool for robot vision,” making it possible to boost performance, versatility, and simplify cell and integration.
Other updates include point cloud painting, Z-range filter
It is not always possible to get a complete point cloud as is required by just capturing from one viewpoint. This means an item will be captured from various viewpoints to get a more complete point cloud.
The challenge here is that the coordinate system of these separate point clouds will differ from each other, because they have been taken from a different viewpoint, either the camera moved, or the position of the item itself changed.
To combat this, the coordinate systems must be transformed into a unified system so that common points are aligned. With the latest update, users can paint an unorganized point cloud with uniform colors, making it easier to see how different point clouds stitch together. The painting feature is also available in the Hand-Eye GUI.
The update includes a Z-range filter slider in Zivid Studio. Zivid said this is useful for simplifying point cloud evaluation. This allows a user to select the depth, or Z-range, that is viewable in Zivid Studio.
As a result, areas of interest in the point cloud can be isolated without the clutter of the entire scene. Previously, the range selection was done via a text box selector. Now this tool can be snapped to the desired range using a mouse to control the range in the depth map window.
Additionally, Ethernet link speed with “connected” status is added in Cameras and Cameras Network configuration dialogs.
Save the date for the 2026 Robotics Summit
