Description

Innovations

AI-Guided 3D Aiming as the Core Innovation: The reading camera has a resolution of 5 megapixels. By sequentially aligning it with each target along the x, y, and z axes using a mirror and liquid lens, a single device covers the entire scene with an effective resolution of more than 190 megapixels, at any distance within the operating range. Rigid camera arrays require numerous cameras distributed across the area, plus a separate row for each depth range.

Predictive Alignment: Using a sequence of overview images, the system determines the three-dimensional trajectory of each target, calculates its future positions, and controls the mirror and focus predictively (patent pending). Instead of blindly scanning predefined positions, it reads specific targets and detects whether a container is missing a label.

In-House High-Speed Control: 3dvisionlabs has developed its own mirror control, trajectory planning, and calibration, including temperature compensation. The ultra-fast mirror moves to the next target within a few milliseconds and continues moving while image capture and processing take place. This enables dozens of targets to be captured per second. For moving objects, the control system tracks the target, compensating for motion blur.

The Innovation Lies in the Integration: Gantry-based solutions assemble cameras, scanners, and industrial PCs from standard components mounted on structural frames. This is the first device to combine scene understanding, 3D target tracking, detailed image capture, and AI computing in a single purpose-built unit, with all components precisely coordinated.

Market relevance

Goods receiving and shipping are critical inspection points in virtually every warehouse and distribution center. By 2027, there will be approximately 214,000 warehouse buildings worldwide measuring more than 4,600 m² (Interact Analysis). Inspections are still often performed manually using handheld scanners or with complex scan gates consisting of numerous individual devices. Missing or incorrectly assigned labels lead to misdeliveries, complaints, and rework.

Target industries include food retail, the automotive and supplier industries, contract logistics, and parcel logistics. Direct customers are system integrators and intralogistics providers that incorporate the camera as a component in conveyor systems, turntables, gates, and inspection stations. By replacing an entire camera array with a single device, the solution makes automated inspection economically viable for medium-sized sites for the first time.

Labor shortages, increasingly stringent traceability requirements, and the trend toward automation are further driving demand. Specific applications are already being designed in collaboration with system integrators.

Market launch is planned for the first quarter of 2027.

Customer benefits

One Device Instead of an Entire Array: Where numerous cameras or scanners are currently required to cover the area, plus additional units for each depth range, a single device is sufficient. For a Euro pallet (1.2 × 2.3 m, 0.2 mm line width, 1.2 m distance), it can theoretically replace around 24 cameras with 12 megapixels each, at a fraction of the hardware cost.

Dramatically Less Installation Effort: No industrial PC, minimal mounting hardware and cabling.

Fast Commissioning: The system is delivered calibrated and configured via an intuitive web interface. Thanks to pre-trained AI functions, no customer-side software development is required. Less work for the system integrator ultimately means lower costs for the end customer.

Complete Inspection: The system reports whether all targets have been read and detects missing labels.

Documentation: Image records of every load unit provide evidence for handling complaints and liability issues.

Reduced Workload: No manual scanning or rechecking of pallets.

Investment Protection: New functions can be added through software updates, and customer-specific AI models can be deployed without replacing the hardware.