Our Product
What is 3D Vision ?
Reference from Zivid.com
3D Vision, often known as a 3D camera, allows three-dimensional information from target objects to be recorded. Laser profiling in conjunction with onboard preprocessing is the most prevalent method. The triangulation principle is used in 3D laser profiling, which involves a camera monitoring a laser line projected onto an object and calculating height information from the deformation of the line profile. Multiple profiles are employed to create a three-dimensional image when the subject passes under the camera. Onboard the camera, as well as on the host or PC system, the 3D information can be calculated.
The target object image in a 3D machine vision system is no longer just a flat image. It’s now a three-dimensional point cloud with accurate coordinates that shows the exact location of every pixel in space. It concurrently offers data in the X, Y, and Z planes, as well as rotational information (around each of the axes).
Our robots can now comfortably handle both shape and position because we’re working with a highly accurate three-dimensional virtual representation of a target object. Regardless of production line climatic circumstances or whether the target object is semi-shiny or light-absorbing dark, they know its precise placement in space, volume, surface angles, degrees of flatness, and characteristics. This makes target item fixturing and overall system design much easier.
As a result of this greatly expanded capabilities, 3D machine vision is now being applied to a wide range of jobs where 2D capability falls short, including, but not limited to:
- Thickness, height and volume measurement
- Dimensioning and space management
- Measuring shapes, holes, angles, and curves
- Detection of surface or assembly defects
- Quality control and verification against 3D CAD models
- Robot guidance and surface tracking (e.g., for welding, gluing, deburring, and more)
- Bin picking for placing, packing or assembly
- Object scanning and digitization
Many people still consider random bin selecting to be a true litmus test for machine vision systems, and it’s easy to see why. Consider the worst-case situation, which involves a bin full of semi-bright, conical steel components. A robot must not only select an individual part from a disorderly pile and determine the optimal grip for it. To avoid collisions and entanglement, it must also take into consideration gripper size, bin sidewalls, and nearby parts at the same time.
To do so, the robot must be able to perceive the parts properly. It needs to know how they’re all lying, their three-dimensional posture, and if they’re totally or partially on top of one another. If they’re dark or light-absorbing, shiny and reflective, they’ll have to deal with shifting scene dynamics like reflections, missing data, and noise. This process must be repeated in a timely and correct manner. The ability to withstand adversity is crucial.
However, using the correct mix of 3D machine vision technology, such complicated, high-accuracy, real-time difficulties can be successfully tackled. Where 2D machine vision fails, it provides efficient, cost-effective alternatives. While 3D machine vision is still a “young child on the block,” its quickly evolving capabilities are light years ahead of their time.
The quality and capabilities of various 3D machine vision technologies might make selecting the proper tool for the job difficult. Because the data produced is significantly more complicated than that provided by a standard camera, it can be quite difficult. More needs, such as resolution, colour, speed, and accuracy, must be carefully considered. Behind the camera, there are other 3D measurement concepts in use. Each task has advantages and disadvantages that must be carefully considered.
The Functions
Assembly automation necessitates consistency, dependability, and speed. It ensures target parts are provided consistently in the process.
3D vision-guided automation
Some machine tending scenarios require inserting or fitting industrial objects like cylinders into a feeding hole or similar, or arrange them in a fixture in a uniform way
Precision motion, pick, and place
This is typically at the input stage of a manufacturing process. A robot is instead emptying a bin bulk-filled with parts for placing on.
Detect and pick randomly stacked parts
The detection and picking of unknown SKUs of various shapes, sizes and materials is still one of the biggest problems in logistics.
Goods to robot, warehouse automation