Object dimensioning
Challenge:
Dimensioning is the process of automated measurement of the width, length, and height of an object (or its bounding box) by means of a visual system. The main industries and applications where dimensioning is used are:
- Transport industry – luggage handling and sorting, load optimization for cargo areas
- Logistics industry – package handling, automated customer billing, pallet dimensioning
- General use – sorting of goods
The main challenges for vision systems used in dimensioning applications include:
- Fast-moving objects
- Objects of different materials and sizes
- High accuracy requirement
- Irregular movement and speed
Conventional 3D vision systems may struggle to provide high enough accuracy at the required speed to capture the whole scanning volume (which limits the object size). The quality of 3D data can be compromised by vibrations or irregular movement – for instance when objects are placed on moving curved conveyors. To achieve satisfactory results, often multiple synchronized vision systems need to be used in parallel.
Solution:
In this particular use case, we deploy two synchronized MotionCam-3Ds to create a 270-degree 3D model of each object that is passing on the conveyor belt at a fast speed. The system then calculates precise dimensions of the objects and this information is further used for the calculation of service costs, for instance, or for optimizing the objects’ placement in a transportation container.
MotionCam-3D provides the best-in-the-market ratio between the quality and speed of data acquisition. It can capture objects moving up to several meters per second while providing a resolution of 0,9 Mpx, submillimeter accuracy of 150 to 900 μm, low noise, detailed contours, and high robustness even when scanning different materials. Varying sizes of the scanned objects pose no challenge – the solution can recognize small (5x5x5 cm) as well as large (150x150x150 cm) objects.
Thanks to these characteristics, the camera enables automated dimensioning of fast-moving objects without interruption or human intervention. It allows precise volume estimation, better cost calculation, faster object processing through higher throughput, and the possibility to sort out oversized objects.
Discussion:
The market offers numerous 3D sensing technologies for scanning moving objects. The most popular ones are Time-of-Flight systems or line scanners. What is the difference and why is the Parallel Structured Light implemented in MotionCam-3D the best option for this application?
Time-of-Flight (ToF) systems are pretty popular as they offer high scanning speed and long scanning range. However, their resolution and accuracy is rather low and the final 3D image is “noisy”. MotionCam-3D, in contrast, provides a high-resolution 3D point cloud with submillimeter accuracy, in addition to its fast scanning speed.
Line scanners provide fast scanning speed too, which makes them suitable for high-speed processing. They use a completely different method of image capturing than area scan vision systems, such as ToF and Parallel Structured Light. Line scanners capture one depth profile at a time in quick succession, for which they require the scanned object or the camera to constantly move. However, their limitation is that they cannot capture random motion and cannot provide such a big scanning range as MotionCam-3D.
Would you like to know more about Photoneo technology and how it can benefit your applications?